Welcome back, everyone! This is Paul McWhorter from TopTechBoy.com, and welcome to Lesson 54 in our AI on the Edge series.

In this lesson, we are taking our hand gesture recognition system to the next level by connecting it directly to real physical hardware! We’re going to take the normalized vector distance calculations and MediaPipe hand-tracking models we’ve been building over the past few lessons, and use them to drive a 12-LED NeoPixel Ring via SPI on the Raspberry Pi in real time.

If you’re following along in the series, pull up a chair, get yourself a nice hot cup of iced coffee, and let’s get right into it!

What We Are Building Today

In previous lessons, we learned how to extract 3D landmark coordinates from MediaPipe, scale them by the palm length to achieve scale-invariant gesture tracking, and store those relative Euclidean distances into a pickle database (gestureNEOPIXEL.pkl).

In Lesson 54, we map those live gesture detections directly to non-blocking LED animation patterns:

Fist: Pulses a dynamic red “charge” sine wave.
Open Palm: Drives the NeoPixel ring at full intensity white beacon.
Peace Sign (✌️): Runs a dual-spinner pattern with opposing green and blue tracer LEDs over a low background red glow.
Thumbs Up: Illumination of a solid, bright green “success” status.
Point: Sweeps a fast green radar chaser dot across a red background array.
Rock / Metal (🤘): Generates a smooth, sinusoidal cyan-to-green energy pulse.
OK Sign: Cycles through a full spectrum RGB rainbow wheel.
Unknown / Default: Gracefully turns off all LEDs.

Hardware Setup & Wiring (SPI Interface)

To ensure high frame rates without flickering, we drive our NeoPixels over the SPI bus using the neopixel_spi library.

Wiring Diagram Guide

NeoPixel VCC: Connect to 5V power supply (ensure adequate power if driving high brightness).
NeoPixel GND: Common ground with the Raspberry Pi GND pin.
NeoPixel DIN (Data In): Connect to GPIO 10 (SPI0 MOSI) on the Raspberry Pi pinout (Physical Pin 19).

Note: Ensure SPI is enabled in your Raspberry Pi system configuration (sudo raspi-config -> Interfaces -> SPI).

⌨️ Interactive Keyboard Controls

While running the main video processing loop, you can dynamically manage your trained gestures directly from your keyboard without interrupting the camera feed:

SPACE: Capture and train/overwrite the current gesture showing on screen.
T: Prompt the terminal to type in and add a brand-new gesture name on the fly.
D: Prompt the terminal to remove/delete an existing gesture from the active dataset.
Q: Cleanly turn off all NeoPixel LEDs, save the updated gesture data, and terminate the program.

Complete Python Source Code

Below is the complete, fully commented Python script for Lesson 54.

# ====================================================================
# DISCLAIMER:
# This code is provided as-is for educational and experimental 
# purposes only. The author makes no representations or warranties of 
# any kind concerning the safety, suitability, or accuracy of this 
# code. Use at your own risk. The author assumes no liability for any 
# damages, system failures, security breaches, or network issues 
# resulting from the use or implementation of this script.
# ====================================================================
import cv2
import time
from picamera2 import Picamera2
import mediapipe as mp
import math
import pickle
import os
import colorsys

# ====================================================================
# NEOPIXEL CONTROL INITIALIZATION (Adafruit neopixel_spi via SPI)
# ====================================================================

import board
import neopixel_spi as neopixel

NUM_LEDS = 12  # Set to match your NeoPixel ring size (e.g., 12)
PO = neopixel.GRB

# Initialize NeoPixel ring using SPI
spi = board.SPI()
strip = neopixel.NeoPixel_SPI(
    spi, 
    NUM_LEDS, 
    pixel_order=PO, 
    auto_write=False
)

# Turn off all LEDs on startup
strip.fill(0)
strip.show()

NEOPIXEL_ENABLED = True
print("[INFO] NeoPixel SPI initialized successfully.")

# Animation state variables (non-blocking updates)
animation_step = 0
lastAnimUpdate = time.time()
# ====================================================================

DATA_FILE = "gestureNEOPIXEL.pkl"

# Function to save the trained gesture dictionary to a file using pickle
def saveTrainedData(filename, data):
    with open(filename, 'wb') as f:
        pickle.dump(data, f)
    print("[INFO] Saved gesture data to " + str(filename))

# Function to load previously saved gesture data from a file
def loadTrainedData(filename):
    if os.path.exists(filename):
        with open(filename, 'rb') as f:
            data = pickle.load(f)
            print("[INFO] Loaded " + str(len(data)) + " gesture(s) from " + str(filename))
        return data
    return {}

# --- SETUP & INITIAL TRAINING ---
trainedData = loadTrainedData(DATA_FILE)
gestureNames = []

if not trainedData:
    print("HAND GESTURE TRAINING SETUP")
    numGestures = int(input("Enter Number of Gestures to Train On: "))
    for idx in range(numGestures):
        name = input("Enter Name for Gesture #" + str(idx + 1) + ": ")
        gestureNames.append(name)

def findDistance(pt1, pt2):
    dX = pt1.x - pt2.x
    dY = pt1.y - pt2.y
    dZ = pt1.z - pt2.z
    return math.sqrt(dX*dX + dY*dY + dZ*dZ)

def getTipDistances(handLm):
    lm = handLm.landmark
    wrist = lm[0]
    indexBase = lm[5]  
    scaleFactor = findDistance(wrist, indexBase)
    tips = [lm[4], lm[8], lm[12], lm[16], lm[20]]
    tipDistances = []
    for tip in tips:
        dist = findDistance(wrist, tip) / scaleFactor
        tipDistances.append(dist)
    for startTip in range(len(tips)):
        for endTip in range(startTip + 1, len(tips)):
            dist = findDistance(tips[startTip], tips[endTip]) / scaleFactor
            tipDistances.append(dist)
    return tipDistances

def calculateError(tipDist, trainDist):
    totalError = 0
    for idx in range(len(tipDist)):
        diff = tipDist[idx] - trainDist[idx]
        totalError += abs(diff)
    return totalError

def findGesture(tipLive, trainedData, maxAllowedError):
    bestMatch = "Unknown"
    lowestError = 999999999
    for gestureName, trainedDist in trainedData.items():
        error = calculateError(tipLive, trainedDist)
        if error < lowestError:
            lowestError = error
            bestMatch = gestureName
    if lowestError > maxAllowedError:
        return "Unknown"
    return bestMatch

# ====================================================================
# NEOPIXEL CONTROL (COLOR HELPER & ANIMATION DISPATCHER)
# ====================================================================
def hsv2rgb(h):
    """ Converts HSV float (0.0-1.0) to RGB integer tuple (0-255) """
    r, g, b = colorsys.hsv_to_rgb(h, 1, 1)
    return (int(r * 255), int(g * 255), int(b * 255))

def updateNeoPixels(gesture):
    """ Non-blocking LED pattern update based on the current detected gesture """
    global animation_step, lastAnimUpdate
    if not NEOPIXEL_ENABLED:
        return

    now = time.time()
    gestureClean = gesture.lower()
    pattern_matched = False

    # 1. FIST / CHARGE (Red pulse)
    if gestureClean == "fist":
        pulse = int(127 + 127 * math.sin(now * 5.0))
        strip.fill((pulse, 0, 0))
        strip.show()
        pattern_matched = True

    # 2. OPEN PALM / BEACON (Solid Bright White)
    if gestureClean == "palm":
        strip.fill((255, 255, 255))
        strip.show()
        pattern_matched = True

    # 3. PEACE SIGN / DUAL SPINNER (Rotating Blue/Cyan dots)
    if gestureClean == "peace":
        if now - lastAnimUpdate > 0.08:
            animation_step = (animation_step + 1) % NUM_LEDS
            lastAnimUpdate = now
        strip.fill((40, 0, 0))  # Background red
        pos1 = animation_step
        pos2 = (animation_step + (NUM_LEDS // 2)) % NUM_LEDS
        strip[pos1] = (0, 255, 0)   # Green spinner dot
        strip[pos2] = (0, 0, 255)   # Blue spinner dot
        strip.show()
        pattern_matched = True

    # 4. THUMBS UP / SUCCESS (Solid Green)
    if gestureClean == "thumbsup":
        strip.fill((0, 255, 0))
        strip.show()
        pattern_matched = True

    # 5. POINTING / CHASER (Radar sweep)
    if gestureClean in "point":
        if now - lastAnimUpdate > 0.06:
            animation_step = (animation_step + 1) % NUM_LEDS
            lastAnimUpdate = now
        strip.fill((255, 0, 0))
        strip[animation_step] = (0, 255, 0)
        strip.show()
        pattern_matched = True

    # 6. ROCK / FIRE (Sinusoidal Green-Cyan pulse)
    if gestureClean == "rock":
        step = int((now * 100) % 360)
        brightness = (math.sin(step * math.pi / 180) + 1) / 2
        g = int(255 * brightness)
        b = int(50 * brightness)
        strip.fill((0, g, b))
        strip.show()
        pattern_matched = True

    # 7. OK SIGN / RAINBOW WHEEL (Rotating RGB spectrum)
    if gestureClean in "ok":
        if now - lastAnimUpdate > 0.03:
            animation_step = (animation_step + 1) % 100
            lastAnimUpdate = now
        hue = animation_step / 100.0
        strip.fill(hsv2rgb(hue))
        strip.show()
        pattern_matched = True

    # DEFAULT / UNKNOWN (LEDs OFF)
    if not pattern_matched:
        strip.fill((0, 0, 0))
        strip.show()
# ====================================================================

W, H = 1280, 720
tStart = time.time()
fps = 15
RES = (W, H)

piCam = Picamera2(1)
piCam.preview_configuration.main.size = RES
piCam.preview_configuration.main.format = "RGB888"
piCam.preview_configuration.controls.FrameRate = 60
piCam.preview_configuration.align()
piCam.configure("preview")
piCam.start()

textLowerLeft = (int(W * .01), int(H * .07))
fontFace = cv2.FONT_HERSHEY_SIMPLEX
fontThickness = int(W / 425)
fontScale = H * .002
fontColor = (0, 0, 255)

textFps = (int(W * .01), int(H * .07))
textGesture = (int(W * .01), int(H * .14))
textGesturePrompt = (int(W * .01), int(H * .21))

maxErrorThreshold = 3.5 

hands = mp.solutions.hands.Hands(
    model_complexity=0,
    min_detection_confidence=.5,
    min_tracking_confidence=.5,
    max_num_hands=2
)

trainIdx = 0

while True:
    deltaT = time.time() - tStart
    tStart = time.time()
    fps = fps * .95 + (1 / deltaT) * .05
    
    frame = piCam.capture_array()
    frame = cv2.flip(frame, -1)
    rgbFrame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    results = hands.process(rgbFrame)
    
    if results.multi_hand_landmarks:
        frameHeight, frameWidth, _ = frame.shape        
        for handLm in results.multi_hand_landmarks:
            for idx, landmark in enumerate(handLm.landmark):
                pixelX = int(landmark.x * frameWidth)
                pixelY = int(landmark.y * frameHeight)
                cv2.putText(
                    frame, str(idx), (pixelX, pixelY),
                    cv2.FONT_HERSHEY_SIMPLEX, .6 * fontScale, (255, 0, 0), fontThickness
                )
                
    if trainIdx < len(gestureNames):
        targetName = gestureNames[trainIdx]
        promptText = "show " + targetName + " & Press Space"
        cv2.putText(frame, promptText, textGesturePrompt, fontFace, fontScale, (0, 255, 255), fontThickness)

    detectGesture = "Unknown"
    if results.multi_hand_landmarks:
        liveDistances = getTipDistances(results.multi_hand_landmarks[0])
        detectGesture = findGesture(liveDistances, trainedData, maxErrorThreshold)

    # ====================================================================
    # EXECUTE HARDWARE PATTERN UPDATE
    # ====================================================================
    updateNeoPixels(detectGesture)
    # ====================================================================

    labelText = "Gesture: " + detectGesture
    cv2.putText(frame, labelText, textGesture, fontFace, fontScale, (0, 255, 0), fontThickness)
    myText = "FPS: " + str(round(fps, 1))
    cv2.putText(frame, myText, textLowerLeft, fontFace, fontScale, fontColor, fontThickness)
    
    cv2.imshow("Camera", frame)
    #cv2.moveWindow("Camera", 0, 60)
    
    key = cv2.waitKey(1)
    
    # Save landmark measurements when Space is pressed
    if key == ord(' '):
        if results.multi_hand_landmarks:
            if trainIdx < len(gestureNames):
                targetHand = results.multi_hand_landmarks[0]
                targetName = gestureNames[trainIdx]
                trainedData[targetName] = getTipDistances(targetHand)
                saveTrainedData(DATA_FILE, trainedData)
                trainIdx += 1

    # 't' key allows adding a NEW gesture dynamically during runtime
    if key == ord('t'):
        newGestureName = input("Enter name for new gesture to train: ").strip()
        if newGestureName:
            gestureNames.append(newGestureName)

    # 'd' key deletes a gesture
    if key == ord('d'):
        delName = input("Enter name of gesture to remove: ").strip()
        if delName in trainedData:
            del trainedData[delName]
            saveTrainedData(DATA_FILE, trainedData)
            print("[INFO] Removed " + delName + " from trained data.")
            if delName in gestureNames:
                gestureNames.remove(delName)
        if delName not in trainedData:
            print("[WARNING] Gesture: " + delName + " not found in trained data.")
        trainIdx=min(trainIdx,len(gestureNames))
    if key == ord('q'):
        break

# Clean up hardware and exit
# ====================================================================
# CLEAR LEDS ON SHUTDOWN
# ====================================================================
if NEOPIXEL_ENABLED:
    strip.fill((0, 0, 0))
    strip.show()
# ====================================================================

cv2.destroyAllWindows()
saveTrainedData(DATA_FILE, trainedData)
print('Program Terminated')

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

# ====================================================================

# DISCLAIMER:

# This code is provided as-is for educational and experimental

# purposes only. The author makes no representations or warranties of

# any kind concerning the safety, suitability, or accuracy of this

# code. Use at your own risk. The author assumes no liability for any

# damages, system failures, security breaches, or network issues

# resulting from the use or implementation of this script.

# ====================================================================

import cv2

import time

from picamera2 import Picamera2

import mediapipe as mp

import math

import pickle

import os

import colorsys

# ====================================================================

# NEOPIXEL CONTROL INITIALIZATION (Adafruit neopixel_spi via SPI)

# ====================================================================

import board

import neopixel_spi as neopixel

NUM_LEDS = 12 # Set to match your NeoPixel ring size (e.g., 12)

PO = neopixel.GRB

# Initialize NeoPixel ring using SPI

spi = board.SPI()

strip = neopixel.NeoPixel_SPI(

spi,

NUM_LEDS,

pixel_order=PO,

auto_write=False

)

# Turn off all LEDs on startup

strip.fill(0)

strip.show()

NEOPIXEL_ENABLED = True

print("[INFO] NeoPixel SPI initialized successfully.")

# Animation state variables (non-blocking updates)

animation_step = 0

lastAnimUpdate = time.time()

# ====================================================================

DATA_FILE = "gestureNEOPIXEL.pkl"

# Function to save the trained gesture dictionary to a file using pickle

def saveTrainedData(filename, data):

with open(filename, 'wb') as f:

pickle.dump(data, f)

print("[INFO] Saved gesture data to " + str(filename))

# Function to load previously saved gesture data from a file

def loadTrainedData(filename):

if os.path.exists(filename):

with open(filename, 'rb') as f:

data = pickle.load(f)

print("[INFO] Loaded " + str(len(data)) + " gesture(s) from " + str(filename))

return data

return {}

# --- SETUP & INITIAL TRAINING ---

trainedData = loadTrainedData(DATA_FILE)

gestureNames = []

if not trainedData:

print("HAND GESTURE TRAINING SETUP")

numGestures = int(input("Enter Number of Gestures to Train On: "))

for idx in range(numGestures):

name = input("Enter Name for Gesture #" + str(idx + 1) + ": ")

gestureNames.append(name)

def findDistance(pt1, pt2):

dX = pt1.x - pt2.x

dY = pt1.y - pt2.y

dZ = pt1.z - pt2.z

return math.sqrt(dX*dX + dY*dY + dZ*dZ)

def getTipDistances(handLm):

lm = handLm.landmark

wrist = lm[0]

indexBase = lm[5]

scaleFactor = findDistance(wrist, indexBase)

tips = [lm[4], lm[8], lm[12], lm[16], lm[20]]

tipDistances = []

for tip in tips:

dist = findDistance(wrist, tip) / scaleFactor

tipDistances.append(dist)

for startTip in range(len(tips)):

for endTip in range(startTip + 1, len(tips)):

dist = findDistance(tips[startTip], tips[endTip]) / scaleFactor

tipDistances.append(dist)

return tipDistances

def calculateError(tipDist, trainDist):

totalError = 0

for idx in range(len(tipDist)):

diff = tipDist[idx] - trainDist[idx]

totalError += abs(diff)

return totalError

def findGesture(tipLive, trainedData, maxAllowedError):

bestMatch = "Unknown"

lowestError = 999999999

for gestureName, trainedDist in trainedData.items():

error = calculateError(tipLive, trainedDist)

if error < lowestError:

lowestError = error

bestMatch = gestureName

if lowestError > maxAllowedError:

return "Unknown"

return bestMatch

# ====================================================================

# NEOPIXEL CONTROL (COLOR HELPER & ANIMATION DISPATCHER)

# ====================================================================

def hsv2rgb(h):

""" Converts HSV float (0.0-1.0) to RGB integer tuple (0-255) """

r, g, b = colorsys.hsv_to_rgb(h, 1, 1)

return (int(r * 255), int(g * 255), int(b * 255))

def updateNeoPixels(gesture):

""" Non-blocking LED pattern update based on the current detected gesture """

global animation_step, lastAnimUpdate

if not NEOPIXEL_ENABLED:

return

now = time.time()

gestureClean = gesture.lower()

pattern_matched = False

# 1. FIST / CHARGE (Red pulse)

if gestureClean == "fist":

pulse = int(127 + 127 * math.sin(now * 5.0))

strip.fill((pulse, 0, 0))

strip.show()

pattern_matched = True

# 2. OPEN PALM / BEACON (Solid Bright White)

if gestureClean == "palm":

strip.fill((255, 255, 255))

strip.show()

pattern_matched = True

# 3. PEACE SIGN / DUAL SPINNER (Rotating Blue/Cyan dots)

if gestureClean == "peace":

if now - lastAnimUpdate > 0.08:

animation_step = (animation_step + 1) % NUM_LEDS

lastAnimUpdate = now

strip.fill((40, 0, 0)) # Background red

pos1 = animation_step

pos2 = (animation_step + (NUM_LEDS // 2)) % NUM_LEDS

strip[pos1] = (0, 255, 0) # Green spinner dot

strip[pos2] = (0, 0, 255) # Blue spinner dot

strip.show()

pattern_matched = True

# 4. THUMBS UP / SUCCESS (Solid Green)

if gestureClean == "thumbsup":

strip.fill((0, 255, 0))

strip.show()

pattern_matched = True

# 5. POINTING / CHASER (Radar sweep)

if gestureClean in "point":

if now - lastAnimUpdate > 0.06:

animation_step = (animation_step + 1) % NUM_LEDS

lastAnimUpdate = now

strip.fill((255, 0, 0))

strip[animation_step] = (0, 255, 0)

strip.show()

pattern_matched = True

# 6. ROCK / FIRE (Sinusoidal Green-Cyan pulse)

if gestureClean == "rock":

step = int((now * 100) % 360)

brightness = (math.sin(step * math.pi / 180) + 1) / 2

g = int(255 * brightness)

b = int(50 * brightness)

strip.fill((0, g, b))

strip.show()

pattern_matched = True

# 7. OK SIGN / RAINBOW WHEEL (Rotating RGB spectrum)

if gestureClean in "ok":

if now - lastAnimUpdate > 0.03:

animation_step = (animation_step + 1) % 100

lastAnimUpdate = now

hue = animation_step / 100.0

strip.fill(hsv2rgb(hue))

strip.show()

pattern_matched = True

# DEFAULT / UNKNOWN (LEDs OFF)

if not pattern_matched:

strip.fill((0, 0, 0))

strip.show()

# ====================================================================

W, H = 1280, 720

tStart = time.time()

fps = 15

RES = (W, H)

piCam = Picamera2(1)

piCam.preview_configuration.main.size = RES

piCam.preview_configuration.main.format = "RGB888"

piCam.preview_configuration.controls.FrameRate = 60

piCam.preview_configuration.align()

piCam.configure("preview")

piCam.start()

textLowerLeft = (int(W * .01), int(H * .07))

fontFace = cv2.FONT_HERSHEY_SIMPLEX

fontThickness = int(W / 425)

fontScale = H * .002

fontColor = (0, 0, 255)

textFps = (int(W * .01), int(H * .07))

textGesture = (int(W * .01), int(H * .14))

textGesturePrompt = (int(W * .01), int(H * .21))

maxErrorThreshold = 3.5

hands = mp.solutions.hands.Hands(

model_complexity=0,

min_detection_confidence=.5,

min_tracking_confidence=.5,

max_num_hands=2

)

trainIdx = 0

while True:

deltaT = time.time() - tStart

tStart = time.time()

fps = fps * .95 + (1 / deltaT) * .05

frame = piCam.capture_array()

frame = cv2.flip(frame, -1)

rgbFrame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

results = hands.process(rgbFrame)

if results.multi_hand_landmarks:

frameHeight, frameWidth, _ = frame.shape

for handLm in results.multi_hand_landmarks:

for idx, landmark in enumerate(handLm.landmark):

pixelX = int(landmark.x * frameWidth)

pixelY = int(landmark.y * frameHeight)

cv2.putText(

frame, str(idx), (pixelX, pixelY),

cv2.FONT_HERSHEY_SIMPLEX, .6 * fontScale, (255, 0, 0), fontThickness

)

if trainIdx < len(gestureNames):

targetName = gestureNames[trainIdx]

promptText = "show " + targetName + " & Press Space"

cv2.putText(frame, promptText, textGesturePrompt, fontFace, fontScale, (0, 255, 255), fontThickness)

detectGesture = "Unknown"

if results.multi_hand_landmarks:

liveDistances = getTipDistances(results.multi_hand_landmarks[0])

detectGesture = findGesture(liveDistances, trainedData, maxErrorThreshold)

# ====================================================================

# EXECUTE HARDWARE PATTERN UPDATE

# ====================================================================

updateNeoPixels(detectGesture)

# ====================================================================

labelText = "Gesture: " + detectGesture

cv2.putText(frame, labelText, textGesture, fontFace, fontScale, (0, 255, 0), fontThickness)

myText = "FPS: " + str(round(fps, 1))

cv2.putText(frame, myText, textLowerLeft, fontFace, fontScale, fontColor, fontThickness)

cv2.imshow("Camera", frame)

#cv2.moveWindow("Camera", 0, 60)

key = cv2.waitKey(1)

# Save landmark measurements when Space is pressed

if key == ord(' '):

if results.multi_hand_landmarks:

if trainIdx < len(gestureNames):

targetHand = results.multi_hand_landmarks[0]

targetName = gestureNames[trainIdx]

trainedData[targetName] = getTipDistances(targetHand)

saveTrainedData(DATA_FILE, trainedData)

trainIdx += 1

# 't' key allows adding a NEW gesture dynamically during runtime

if key == ord('t'):

newGestureName = input("Enter name for new gesture to train: ").strip()

if newGestureName:

gestureNames.append(newGestureName)

# 'd' key deletes a gesture

if key == ord('d'):

delName = input("Enter name of gesture to remove: ").strip()

if delName in trainedData:

del trainedData[delName]

saveTrainedData(DATA_FILE, trainedData)

print("[INFO] Removed " + delName + " from trained data.")

if delName in gestureNames:

gestureNames.remove(delName)

if delName not in trainedData:

print("[WARNING] Gesture: " + delName + " not found in trained data.")

trainIdx=min(trainIdx,len(gestureNames))

if key == ord('q'):

break

# Clean up hardware and exit

# ====================================================================

# CLEAR LEDS ON SHUTDOWN

# ====================================================================

if NEOPIXEL_ENABLED:

strip.fill((0, 0, 0))

strip.show()

# ====================================================================

cv2.destroyAllWindows()

saveTrainedData(DATA_FILE, trainedData)

print('Program Terminated')

Homework Assignment

Your homework for this lesson is to modify the updateNeoPixels() function to add two new custom hand gestures of your own creation! Try implementing custom animations like a warm candle flicker or a pulsating dual-color pattern. Leave a link to your solution videos in the comments below!

AI On the Edge, Raspberry Pi

AI on the Edge LESSON 53: Gesture Recognition on the Pi 5 with Training File

October 4, 2026 admin

Welcome back, everyone! In this lesson, we are taking our custom MediaPipe hand gesture recognition system to the next level. Up until now, every time you stopped your Python script, all of your hard-earned gesture training data vanished into thin air. If you wanted to run your project again, you had to re-train every single hand sign from scratch.

That ends today. We are going to make our AI remember what it learned by saving our gesture feature vectors to a local file using Python’s pickle library.

What You Will Learn in This Lesson

Data Persistence: How to serialize and save Python dictionaries containing landmark distance vectors directly to disk (gesture_data.pkl).
Automated Setup: How to check for existing training files at boot so your program automatically skips the initial training prompt if data already exists.
Dynamic Training Management: How to interactively train new gestures on the fly ('t' key) or delete outdated gestures ('d' key) during runtime while automatically updating the saved file.
Robust Error Handling: Ensuring index tracking (trainIdx) safely syncs with live gesture arrays so pressing the spacebar never crashes your video feed.

Key Concepts Explained

1. Pickling Data Objects In MediaPipe gesture recognition, a trained gesture isn’t a complex neural network weight file; it’s simply a dictionary mapping a gesture name (string) to a list of normalized keypoint ratios (floating-point numbers). Python’s pickle module allows us to convert this dictionary into a byte stream and write it directly to a file (.pkl). When the program launches, pickle.load() reconstructs the dictionary in milliseconds.

2. Managing Execution Flow on Startup When gesture_data.pkl exists, our program automatically populates gestureNames with the keys from trainedData and sets trainIdx past the end of the array. This keeps the camera pipeline running in Inference Mode immediately. If no file exists, the script seamlessly drops into Training Setup Mode to prompt the user for initial inputs.

This is the code we develop in this video lesson:

# ====================================================================
# DISCLAIMER:
# This code is provided as-is for educational and experimental 
# purposes only. The author makes no representations or warranties of 
# any kind concerning the safety, suitability, or accuracy of this 
# code. Use at your own risk. The author assumes no liability for any 
# damages, system failures, security breaches, or network issues 
# resulting from the use or implementation of this script.
# ====================================================================
import cv2
import time
from picamera2 import Picamera2
import mediapipe as mp
import math
import os
import pickle

DATA_FILE = "gesture_data.pkl"

def saveTrainedData(filename, data):
    with open(filename, 'wb') as f:
        pickle.dump(data,f)
    print("[INFO] Saved Gesture Data to "+str(filename))
def loadTrainedData(filename):
    if os.path.exists(filename):
        with open(filename, 'rb') as f:
            data = pickle.load(f)
            print("[INFO] LOADED "+str(len(data))+ " gestures from " + str(filename))
            return data
    if not os.path.exists(filename):
        return {}

def findDistance(pt1, pt2):
    dX = pt1.x - pt2.x
    dY = pt1.y - pt2.y
    dZ = pt1.z - pt2.z
    dist = math.sqrt(dX*dX + dY*dY + dZ*dZ)
    return dist

def getTipDistances(handLm):
    lm = handLm.landmark
    wrist = lm[0]
    indexBase = lm[5]  
    scaleFactor = findDistance(wrist, indexBase)
    tips = [lm[4],lm[8],lm[12],lm[16],lm[20]]
    tipDistances =[]
    for tip in tips:
        dist = findDistance( wrist, tip) / scaleFactor
        tipDistances.append(dist)
    for startTip in range(len(tips)):
        for endTip in range(startTip+1, len(tips)):
            dist = findDistance(tips[startTip],tips[endTip]) / scaleFactor
            tipDistances.append(dist)
    return tipDistances

def calculateError(tipDist, trainDist):
    totalError = 0
    for idx in range(len(tipDist)):
        diff = tipDist[idx] -trainDist[idx]
        totalError = totalError + abs(diff)
    return totalError

def findGesture(tipLive, trainedData, maxAllowedError):
    bestMatch = "Unknown"
    lowestError = 999999999
    for gestureName, trainedDist in trainedData.items():
        error  = calculateError(tipLive, trainedDist)
        if error < lowestError:
            lowestError = error
            bestMatch = gestureName
    if lowestError > maxAllowedError:
        return "Unknown"
    if lowestError <= maxAllowedError:
        return bestMatch
trainedData = loadTrainedData(DATA_FILE)
gestureNames =[]
if not trainedData:
    print("HAND GESTURE TRAINING SETUP")
    numGestures = int(input("Enter Number of Gestures to Train On: "))
    gestureNames =[]
    for idx in range(numGestures):
        name = input("Enter Name for Gesture #"+ str(idx+1)+ ": ")
        gestureNames.append(name)
W=1280
H=720
tStart = time.time()
fps = 15
RES = (W,H)
piCam = Picamera2(1)
piCam.preview_configuration.main.size = RES
piCam.preview_configuration.main.format = "RGB888"
piCam.preview_configuration.controls.FrameRate=60
piCam.preview_configuration.align()
piCam.configure("preview")
piCam.start()

textLowerLeft = (int(W*.01),int(H*.07))
fontFace = cv2.FONT_HERSHEY_SIMPLEX
fontThickness = int(W/425)
fontScale = H*.002
fontColor = (0,0,255)

textFps = (int(W*.01),int(H*.07))
textGesture= (int(W*.01),int(H*.14))
textGesturePrompt = (int(W*.01),int(H*.21))

maxErrorThreshold = 3.5 

hands = mp.solutions.hands.Hands(
    model_complexity =  0,
    min_detection_confidence = .5,
    min_tracking_confidence = .5,
    max_num_hands =2)
trainIdx = 0

while True:
    deltaT = time.time() - tStart
    tStart=time.time()
    fps = fps*.95 + (1/deltaT)*.05
    frame= piCam.capture_array()
    frame=cv2.flip(frame,-1)
    rgbFrame= cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    results = hands.process(rgbFrame)
    if results.multi_hand_landmarks:
        frameHeight, frameWidth, _ = frame.shape        
        for handLm in results.multi_hand_landmarks:
            # Draw landmark numbers on hand joints
            for idx, landmark in enumerate(handLm.landmark):
                pixelX = int(landmark.x * frameWidth)
                pixelY = int(landmark.y * frameHeight)
                cv2.putText(
                    frame, str(idx), (pixelX, pixelY),
                    cv2.FONT_HERSHEY_SIMPLEX, .6*fontScale, (255, 0, 0), fontThickness
                )
    if trainIdx < len(gestureNames):
        targetName = gestureNames[trainIdx]
        promptText = "show " + targetName + " & Press Space"
        cv2.putText(frame,promptText, textGesturePrompt, fontFace,fontScale, (0,255,255), fontThickness)
    detectGesture = "Unknown"
    if results.multi_hand_landmarks:
        liveDistances = getTipDistances(results.multi_hand_landmarks[0])
        detectGesture = findGesture(liveDistances, trainedData, maxErrorThreshold)
    labelText = "Gesture: " + detectGesture
    cv2.putText(frame, labelText, textGesture,fontFace,fontScale, (0,255,0),fontThickness)
    myText = "FPS: "+str(round(fps,1))
    cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)
    cv2.imshow("Camera", frame)
    #cv2.moveWindow("Camera",0,60)
    key = cv2.waitKey(1)
    if key == ord('t'):
        newGestureName = input("Enter Name for New Gesture: ")
        gestureNames.append(newGestureName)
        saveTrainedData(DATA_FILE, trainedData)
    if key == ord('d'):
        delName = input("Enter Name of Gesture to Remove: ")
        if delName not in trainedData:
            print("[WARNING] Gesture "+delName+ " Not Found in Training Data.")
        if delName in trainedData:
            del trainedData[delName]
            saveTrainedData(DATA_FILE,trainedData)
            print("[INFO] Removed "+delName+"from trained data.")
            if delName in gestureNames:
                gestureNames.remove(delName)
            trainIdx=min(trainIdx,len(gestureNames))
    if key == ord(' '):
        if results.multi_hand_landmarks:
            targetHand = results.multi_hand_landmarks[0]
            trainedData[targetName] = getTipDistances(targetHand)
            trainIdx = trainIdx + 1
    if key == ord('q'):
        break
cv2.destroyAllWindows()
saveTrainedData(DATA_FILE, trainedData)
print('Program Terminated')

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

# ====================================================================

# DISCLAIMER:

# This code is provided as-is for educational and experimental

# purposes only. The author makes no representations or warranties of

# any kind concerning the safety, suitability, or accuracy of this

# code. Use at your own risk. The author assumes no liability for any

# damages, system failures, security breaches, or network issues

# resulting from the use or implementation of this script.

# ====================================================================

import cv2

import time

from picamera2 import Picamera2

import mediapipe as mp

import math

import os

import pickle

DATA_FILE = "gesture_data.pkl"

def saveTrainedData(filename, data):

with open(filename, 'wb') as f:

pickle.dump(data,f)

print("[INFO] Saved Gesture Data to "+str(filename))

def loadTrainedData(filename):

if os.path.exists(filename):

with open(filename, 'rb') as f:

data = pickle.load(f)

print("[INFO] LOADED "+str(len(data))+ " gestures from " + str(filename))

return data

if not os.path.exists(filename):

return {}

def findDistance(pt1, pt2):

dX = pt1.x - pt2.x

dY = pt1.y - pt2.y

dZ = pt1.z - pt2.z

dist = math.sqrt(dX*dX + dY*dY + dZ*dZ)

return dist

def getTipDistances(handLm):

lm = handLm.landmark

wrist = lm[0]

indexBase = lm[5]

scaleFactor = findDistance(wrist, indexBase)

tips = [lm[4],lm[8],lm[12],lm[16],lm[20]]

tipDistances =[]

for tip in tips:

dist = findDistance( wrist, tip) / scaleFactor

tipDistances.append(dist)

for startTip in range(len(tips)):

for endTip in range(startTip+1, len(tips)):

dist = findDistance(tips[startTip],tips[endTip]) / scaleFactor

tipDistances.append(dist)

return tipDistances

def calculateError(tipDist, trainDist):

totalError = 0

for idx in range(len(tipDist)):

diff = tipDist[idx] -trainDist[idx]

totalError = totalError + abs(diff)

return totalError

def findGesture(tipLive, trainedData, maxAllowedError):

bestMatch = "Unknown"

lowestError = 999999999

for gestureName, trainedDist in trainedData.items():

error = calculateError(tipLive, trainedDist)

if error < lowestError:

lowestError = error

bestMatch = gestureName

if lowestError > maxAllowedError:

return "Unknown"

if lowestError <= maxAllowedError:

return bestMatch

trainedData = loadTrainedData(DATA_FILE)

gestureNames =[]

if not trainedData:

print("HAND GESTURE TRAINING SETUP")

numGestures = int(input("Enter Number of Gestures to Train On: "))

gestureNames =[]

for idx in range(numGestures):

name = input("Enter Name for Gesture #"+ str(idx+1)+ ": ")

gestureNames.append(name)

W=1280

H=720

tStart = time.time()

fps = 15

RES = (W,H)

piCam = Picamera2(1)

piCam.preview_configuration.main.size = RES

piCam.preview_configuration.main.format = "RGB888"

piCam.preview_configuration.controls.FrameRate=60

piCam.preview_configuration.align()

piCam.configure("preview")

piCam.start()

textLowerLeft = (int(W*.01),int(H*.07))

fontFace = cv2.FONT_HERSHEY_SIMPLEX

fontThickness = int(W/425)

fontScale = H*.002

fontColor = (0,0,255)

textFps = (int(W*.01),int(H*.07))

textGesture= (int(W*.01),int(H*.14))

textGesturePrompt = (int(W*.01),int(H*.21))

maxErrorThreshold = 3.5

hands = mp.solutions.hands.Hands(

model_complexity = 0,

min_detection_confidence = .5,

min_tracking_confidence = .5,

max_num_hands =2)

trainIdx = 0

while True:

deltaT = time.time() - tStart

tStart=time.time()

fps = fps*.95 + (1/deltaT)*.05

frame= piCam.capture_array()

frame=cv2.flip(frame,-1)

rgbFrame= cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

results = hands.process(rgbFrame)

if results.multi_hand_landmarks:

frameHeight, frameWidth, _ = frame.shape

for handLm in results.multi_hand_landmarks:

# Draw landmark numbers on hand joints

for idx, landmark in enumerate(handLm.landmark):

pixelX = int(landmark.x * frameWidth)

pixelY = int(landmark.y * frameHeight)

cv2.putText(

frame, str(idx), (pixelX, pixelY),

cv2.FONT_HERSHEY_SIMPLEX, .6*fontScale, (255, 0, 0), fontThickness

)

if trainIdx < len(gestureNames):

targetName = gestureNames[trainIdx]

promptText = "show " + targetName + " & Press Space"

cv2.putText(frame,promptText, textGesturePrompt, fontFace,fontScale, (0,255,255), fontThickness)

detectGesture = "Unknown"

if results.multi_hand_landmarks:

liveDistances = getTipDistances(results.multi_hand_landmarks[0])

detectGesture = findGesture(liveDistances, trainedData, maxErrorThreshold)

labelText = "Gesture: " + detectGesture

cv2.putText(frame, labelText, textGesture,fontFace,fontScale, (0,255,0),fontThickness)

myText = "FPS: "+str(round(fps,1))

cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)

cv2.imshow("Camera", frame)

#cv2.moveWindow("Camera",0,60)

key = cv2.waitKey(1)

if key == ord('t'):

newGestureName = input("Enter Name for New Gesture: ")

gestureNames.append(newGestureName)

saveTrainedData(DATA_FILE, trainedData)

if key == ord('d'):

delName = input("Enter Name of Gesture to Remove: ")

if delName not in trainedData:

print("[WARNING] Gesture "+delName+ " Not Found in Training Data.")

if delName in trainedData:

del trainedData[delName]

saveTrainedData(DATA_FILE,trainedData)

print("[INFO] Removed "+delName+"from trained data.")

if delName in gestureNames:

gestureNames.remove(delName)

trainIdx=min(trainIdx,len(gestureNames))

if key == ord(' '):

if results.multi_hand_landmarks:

targetHand = results.multi_hand_landmarks[0]

trainedData[targetName] = getTipDistances(targetHand)

trainIdx = trainIdx + 1

if key == ord('q'):

break

cv2.destroyAllWindows()

saveTrainedData(DATA_FILE, trainedData)

print('Program Terminated')

Homework Assignment

Now that your Pi 5 can remember hand gestures across reboots, it’s time to bridge software recognition with real-world hardware control!

The Challenge: Integrate an SPI NeoPixel Ring (WS2812B/SK6812) with your gesture recognition code.

Map distinct hand signs to dynamic LED patterns (for example: Fist = Red Pulse, Open Palm = Bright White Beacon, Peace Sign = Dual Spinner, OK Sign = Rainbow Wheel).
Make sure your animation updates are smooth and non-blocking so your OpenCV camera frame rate doesn’t drop!

We want to see who in the community can build the most impressive physical light demonstration controlled entirely by hand gestures. Share your video results and code modifications in the comments below!

AI On the Edge, Raspberry Pi

AI on the Edge LESSON 52: Gesture Recognition on the Raspberry Pi 5

September 28, 2026 admin

Welcome back, everyone! In today’s lesson, AI on the Edge LESSON 52: Gesture Recognition on the Raspberry Pi 5, we are building a custom, real-time hand gesture recognition system from scratch. Up until now, we have been using MediaPipe to track keypoints, identify individual hand joints, and extract raw 3D coordinate data. But raw coordinates alone don’t tell the computer what your hand is actually doing. Today, we bridge that gap by teaching our Raspberry Pi 5 how to train on custom hand positions and dynamically classify live hand gestures using pure mathematics and vector geometry—no heavy neural network retraining required.

The core secret to making gesture recognition work reliably across different camera distances is scale-invariant distance normalization. If you simply measure the pixel distance between your fingertips and your wrist, bringing your hand closer to the camera lens will blow out those numbers and break your classifier. To solve this, our script establishes a baseline unit of measurement for every single frame: the distance between the wrist (landmark 0) and the index finger base joint (landmark 5). By dividing all measured finger-tip-to-wrist and tip-to-tip distances by this dynamic scale factor, the resulting feature vector remains identical whether your hand is two feet away or five feet away from the lens.

Once we extract this normalized 15-element feature vector—representing the spatial ratios between all five fingertips relative to the wrist and each other—we construct an interactive training phase directly inside the live video loop. The program prompts you in the terminal for the number of custom gestures you wish to record, along with their labels (such as “Peace”, “Thumbs Up”, or “Fist”). As you present each gesture to your camera feed and press the spacebar, the algorithm captures the instantaneous geometric signature of your hand and stores it in a runtime training dictionary.

During live inference, our classifier calculates the Sum of Absolute Errors (SAE), or Manhattan distance, between the live incoming feature vector and every stored profile in our training dataset. The function loops through all saved gestures to find the candidate with the absolute lowest cumulative error score. To prevent false positives when an untrained or messy hand pose is shown, we compare that lowest error score against a strict threshold value (maxErrorThreshold = 3.5). If the error falls within the allowable boundary, the matched gesture label is instantly displayed on the live OpenCV HUD; otherwise, the system safely defaults to “Unknown”.

We manage our camera feed using high-speed frame capture with Picamera2 running at 60 frames per second at 1280×720 resolution, processing hand detection with MediaPipe Hands, and rendering real-time performance diagnostics—including an exponential moving average FPS counter—directly onto the video output. Go ahead, study the methodology, load the concept onto your Raspberry Pi 5, grab a hot cup of coffee, and let’s get building!

# ====================================================================
# DISCLAIMER:
# This code is provided as-is for educational and experimental 
# purposes only. The author makes no representations or warranties of 
# any kind concerning the safety, suitability, or accuracy of this 
# code. Use at your own risk. The author assumes no liability for any 
# damages, system failures, security breaches, or network issues 
# resulting from the use or implementation of this script.
# ====================================================================
import cv2
import time
from picamera2 import Picamera2
import mediapipe as mp
import math

def findDistance(pt1, pt2):
    dX = pt1.x - pt2.x
    dY = pt1.y - pt2.y
    dZ = pt1.z - pt2.z
    dist = math.sqrt(dX*dX + dY*dY + dZ*dZ)
    return dist

def getTipDistances(handLm):
    lm = handLm.landmark
    wrist = lm[0]
    indexBase = lm[5]  
    scaleFactor = findDistance(wrist, indexBase)
    tips = [lm[4],lm[8],lm[12],lm[16],lm[20]]
    tipDistances =[]
    for tip in tips:
        dist = findDistance( wrist, tip) / scaleFactor
        tipDistances.append(dist)
    for startTip in range(len(tips)):
        for endTip in range(startTip+1, len(tips)):
            dist = findDistance(tips[startTip],tips[endTip]) / scaleFactor
            tipDistances.append(dist)
    return tipDistances

def calculateError(tipDist, trainDist):
    totalError = 0
    for idx in range(len(tipDist)):
        diff = tipDist[idx] -trainDist[idx]
        totalError = totalError + abs(diff)
    return totalError

def findGesture(tipLive, trainedData, maxAllowedError):
    bestMatch = "Unknown"
    lowestError = 999999999
    for gestureName, trainedDist in trainedData.items():
        error  = calculateError(tipLive, trainedDist)
        if error < lowestError:
            lowestError = error
            bestMatch = gestureName
    if lowestError > maxAllowedError:
        return "Unknown"
    if lowestError <= maxAllowedError:
        return bestMatch
print("HAND GESTURE TRAINING SETUP")
numGestures = int(input("Enter Number of Gestures to Train On: "))
gestureNames =[]
for idx in range(numGestures):
    name = input("Enter Name for Gesture #"+ str(idx+1)+ ": ")
    gestureNames.append(name)
W=1280
H=720
tStart = time.time()
fps = 15
RES = (W,H)
piCam = Picamera2(1)
piCam.preview_configuration.main.size = RES
piCam.preview_configuration.main.format = "RGB888"
piCam.preview_configuration.controls.FrameRate=60
piCam.preview_configuration.align()
piCam.configure("preview")
piCam.start()

textLowerLeft = (int(W*.01),int(H*.07))
fontFace = cv2.FONT_HERSHEY_SIMPLEX
fontThickness = int(W/425)
fontScale = H*.002
fontColor = (0,0,255)

textFps = (int(W*.01),int(H*.07))
textGesture= (int(W*.01),int(H*.14))
textGesturePrompt = (int(W*.01),int(H*.21))

maxErrorThreshold = 3.5 

hands = mp.solutions.hands.Hands(
    model_complexity =  0,
    min_detection_confidence = .5,
    min_tracking_confidence = .5,
    max_num_hands =2)

trainedData ={}
trainIdx = 0

while True:
    deltaT = time.time() - tStart
    tStart=time.time()
    fps = fps*.95 + (1/deltaT)*.05
    frame= piCam.capture_array()
    frame=cv2.flip(frame,-1)
    rgbFrame= cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    results = hands.process(rgbFrame)
    if results.multi_hand_landmarks:
        frameHeight, frameWidth, _ = frame.shape        
        for handLm in results.multi_hand_landmarks:
            # Draw landmark numbers on hand joints
            for idx, landmark in enumerate(handLm.landmark):
                pixelX = int(landmark.x * frameWidth)
                pixelY = int(landmark.y * frameHeight)
                cv2.putText(
                    frame, str(idx), (pixelX, pixelY),
                    cv2.FONT_HERSHEY_SIMPLEX, .6*fontScale, (255, 0, 0), fontThickness
                )
    if trainIdx < len(gestureNames):
        targetName = gestureNames[trainIdx]
        promptText = "show " + targetName + " & Press Space"
        cv2.putText(frame,promptText, textGesturePrompt, fontFace,fontScale, (0,255,255), fontThickness)
        key = cv2.waitKey(1)
        if key == ord(' '):
            if results.multi_hand_landmarks:
                targetHand = results.multi_hand_landmarks[0]
                trainedData[targetName] = getTipDistances(targetHand)
                trainIdx = trainIdx + 1
        if key == ord('q'):
            break
    detectGesture = "Unknown"
    if results.multi_hand_landmarks:
        liveDistances = getTipDistances(results.multi_hand_landmarks[0])
        detectGesture = findGesture(liveDistances, trainedData, maxErrorThreshold)
    labelText = "Gesture: " + detectGesture
    cv2.putText(frame, labelText, textGesture,fontFace,fontScale, (0,255,0),fontThickness)
    myText = "FPS: "+str(round(fps,1))
    cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)
    cv2.imshow("Camera", frame)
    cv2.moveWindow("Camera",0,60)
    if cv2.waitKey(1)==ord('q'):
        break
cv2.destroyAllWindows()
print('Program Terminated')

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

# ====================================================================

# DISCLAIMER:

# This code is provided as-is for educational and experimental

# purposes only. The author makes no representations or warranties of

# any kind concerning the safety, suitability, or accuracy of this

# code. Use at your own risk. The author assumes no liability for any

# damages, system failures, security breaches, or network issues

# resulting from the use or implementation of this script.

# ====================================================================

import cv2

import time

from picamera2 import Picamera2

import mediapipe as mp

import math

def findDistance(pt1, pt2):

dX = pt1.x - pt2.x

dY = pt1.y - pt2.y

dZ = pt1.z - pt2.z

dist = math.sqrt(dX*dX + dY*dY + dZ*dZ)

return dist

def getTipDistances(handLm):

lm = handLm.landmark

wrist = lm[0]

indexBase = lm[5]

scaleFactor = findDistance(wrist, indexBase)

tips = [lm[4],lm[8],lm[12],lm[16],lm[20]]

tipDistances =[]

for tip in tips:

dist = findDistance( wrist, tip) / scaleFactor

tipDistances.append(dist)

for startTip in range(len(tips)):

for endTip in range(startTip+1, len(tips)):

dist = findDistance(tips[startTip],tips[endTip]) / scaleFactor

tipDistances.append(dist)

return tipDistances

def calculateError(tipDist, trainDist):

totalError = 0

for idx in range(len(tipDist)):

diff = tipDist[idx] -trainDist[idx]

totalError = totalError + abs(diff)

return totalError

def findGesture(tipLive, trainedData, maxAllowedError):

bestMatch = "Unknown"

lowestError = 999999999

for gestureName, trainedDist in trainedData.items():

error = calculateError(tipLive, trainedDist)

if error < lowestError:

lowestError = error

bestMatch = gestureName

if lowestError > maxAllowedError:

return "Unknown"

if lowestError <= maxAllowedError:

return bestMatch

print("HAND GESTURE TRAINING SETUP")

numGestures = int(input("Enter Number of Gestures to Train On: "))

gestureNames =[]

for idx in range(numGestures):

name = input("Enter Name for Gesture #"+ str(idx+1)+ ": ")

gestureNames.append(name)

W=1280

H=720

tStart = time.time()

fps = 15

RES = (W,H)

piCam = Picamera2(1)

piCam.preview_configuration.main.size = RES

piCam.preview_configuration.main.format = "RGB888"

piCam.preview_configuration.controls.FrameRate=60

piCam.preview_configuration.align()

piCam.configure("preview")

piCam.start()

textLowerLeft = (int(W*.01),int(H*.07))

fontFace = cv2.FONT_HERSHEY_SIMPLEX

fontThickness = int(W/425)

fontScale = H*.002

fontColor = (0,0,255)

textFps = (int(W*.01),int(H*.07))

textGesture= (int(W*.01),int(H*.14))

textGesturePrompt = (int(W*.01),int(H*.21))

maxErrorThreshold = 3.5

hands = mp.solutions.hands.Hands(

model_complexity = 0,

min_detection_confidence = .5,

min_tracking_confidence = .5,

max_num_hands =2)

trainedData ={}

trainIdx = 0

while True:

deltaT = time.time() - tStart

tStart=time.time()

fps = fps*.95 + (1/deltaT)*.05

frame= piCam.capture_array()

frame=cv2.flip(frame,-1)

rgbFrame= cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

results = hands.process(rgbFrame)

if results.multi_hand_landmarks:

frameHeight, frameWidth, _ = frame.shape

for handLm in results.multi_hand_landmarks:

# Draw landmark numbers on hand joints

for idx, landmark in enumerate(handLm.landmark):

pixelX = int(landmark.x * frameWidth)

pixelY = int(landmark.y * frameHeight)

cv2.putText(

frame, str(idx), (pixelX, pixelY),

cv2.FONT_HERSHEY_SIMPLEX, .6*fontScale, (255, 0, 0), fontThickness

)

if trainIdx < len(gestureNames):

targetName = gestureNames[trainIdx]

promptText = "show " + targetName + " & Press Space"

cv2.putText(frame,promptText, textGesturePrompt, fontFace,fontScale, (0,255,255), fontThickness)

key = cv2.waitKey(1)

if key == ord(' '):

if results.multi_hand_landmarks:

targetHand = results.multi_hand_landmarks[0]

trainedData[targetName] = getTipDistances(targetHand)

trainIdx = trainIdx + 1

if key == ord('q'):

break

detectGesture = "Unknown"

if results.multi_hand_landmarks:

liveDistances = getTipDistances(results.multi_hand_landmarks[0])

detectGesture = findGesture(liveDistances, trainedData, maxErrorThreshold)

labelText = "Gesture: " + detectGesture

cv2.putText(frame, labelText, textGesture,fontFace,fontScale, (0,255,0),fontThickness)

myText = "FPS: "+str(round(fps,1))

cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)

cv2.imshow("Camera", frame)

cv2.moveWindow("Camera",0,60)

if cv2.waitKey(1)==ord('q'):

break

cv2.destroyAllWindows()

print('Program Terminated')

AI On the Edge, Raspberry Pi

Hand Recognition on the Raspberry Pi 5 With openCV and Mediapipe

September 25, 2026 admin

This is a base program which allows you to use mediapipe on the Raspberry Pi 5 under openCV to identify each landmark on a hand. This program is important as we can see the index of each landmark. This will serve as the basis for our next program which will allow us to begin to do gesture recogntion.

# ====================================================================
# DISCLAIMER:
# This code is provided as-is for educational and experimental 
# purposes only. The author makes no representations or warranties of 
# any kind concerning the safety, suitability, or accuracy of this 
# code. Use at your own risk. The author assumes no liability for any 
# damages, system failures, security breaches, or network issues 
# resulting from the use or implementation of this script.
# ====================================================================
import cv2
import time
from picamera2 import Picamera2
import mediapipe as mp

W=1280
H=720
tStart = time.time()
fps = 15
RES = (W,H)
piCam = Picamera2(1)
piCam.preview_configuration.main.size = RES
piCam.preview_configuration.main.format = "RGB888"
piCam.preview_configuration.controls.FrameRate=60
piCam.preview_configuration.align()
piCam.configure("preview")
piCam.start()

textLowerLeft = (int(W*.01),int(H*.07))
fontFace = cv2.FONT_HERSHEY_SIMPLEX
fontThickness = int(W/425)
fontScale = H*.002
fontColor = (0,0,255)

hands = mp.solutions.hands.Hands(
    model_complexity =  0,
    min_detection_confidence = .5,
    min_tracking_confidence = .5,
    max_num_hands =2)

while True:
    deltaT = time.time() - tStart
    tStart=time.time()
    fps = fps*.95 + (1/deltaT)*.05
    frame= piCam.capture_array()
    frame=cv2.flip(frame,-1)
    rgbFrame= cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    results = hands.process(rgbFrame)
    if results.multi_hand_landmarks:
        frameHeight, frameWidth, _ = frame.shape        
        for handLm in results.multi_hand_landmarks:
            # Draw landmark numbers on hand joints
            for idx, landmark in enumerate(handLm.landmark):
                pixelX = int(landmark.x * frameWidth)
                pixelY = int(landmark.y * frameHeight)
                cv2.putText(
                    frame, str(idx), (pixelX, pixelY),
                    cv2.FONT_HERSHEY_SIMPLEX, .6*fontScale, (255, 0, 0), fontThickness
                )
    myText = "FPS: "+str(round(fps,1))
    cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)
    cv2.imshow("Camera", frame)
    cv2.moveWindow("Camera",0,60)
    if cv2.waitKey(1)==ord('q'):
        break
cv2.destroyAllWindows()
print('Program Terminated')

# ====================================================================

# DISCLAIMER:

# This code is provided as-is for educational and experimental

# purposes only. The author makes no representations or warranties of

# any kind concerning the safety, suitability, or accuracy of this

# code. Use at your own risk. The author assumes no liability for any

# damages, system failures, security breaches, or network issues

# resulting from the use or implementation of this script.

# ====================================================================

import cv2

import time

from picamera2 import Picamera2

import mediapipe as mp

W=1280

H=720

tStart = time.time()

fps = 15

RES = (W,H)

piCam = Picamera2(1)

piCam.preview_configuration.main.size = RES

piCam.preview_configuration.main.format = "RGB888"

piCam.preview_configuration.controls.FrameRate=60

piCam.preview_configuration.align()

piCam.configure("preview")

piCam.start()

textLowerLeft = (int(W*.01),int(H*.07))

fontFace = cv2.FONT_HERSHEY_SIMPLEX

fontThickness = int(W/425)

fontScale = H*.002

fontColor = (0,0,255)

hands = mp.solutions.hands.Hands(

model_complexity = 0,

min_detection_confidence = .5,

min_tracking_confidence = .5,

max_num_hands =2)

while True:

deltaT = time.time() - tStart

tStart=time.time()

fps = fps*.95 + (1/deltaT)*.05

frame= piCam.capture_array()

frame=cv2.flip(frame,-1)

rgbFrame= cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

results = hands.process(rgbFrame)

if results.multi_hand_landmarks:

frameHeight, frameWidth, _ = frame.shape

for handLm in results.multi_hand_landmarks:

# Draw landmark numbers on hand joints

for idx, landmark in enumerate(handLm.landmark):

pixelX = int(landmark.x * frameWidth)

pixelY = int(landmark.y * frameHeight)

cv2.putText(

frame, str(idx), (pixelX, pixelY),

cv2.FONT_HERSHEY_SIMPLEX, .6*fontScale, (255, 0, 0), fontThickness

)

myText = "FPS: "+str(round(fps,1))

cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)

cv2.imshow("Camera", frame)

cv2.moveWindow("Camera",0,60)

if cv2.waitKey(1)==ord('q'):

break

cv2.destroyAllWindows()

print('Program Terminated')

AI On the Edge

AI on the Edge LESSON 51: Gesture Control of NeoPixel Ring

September 18, 2026 admin

Welcome back, everyone! In today’s class, AI on the Edge LESSON 51: Gesture Control of NeoPixel Ring, we are bridging the gap between computer vision and physical hardware in a big way. Up until now, we have spent a lot of time tracking hands, plotting 3D landmarks, and rendering graphical user interfaces directly onto our camera feeds. But today, we take those virtual interactions and project them out into the real physical world using a 12-LED NeoPixel ring driven over SPI by our Raspberry Pi 5.

The core concept of this lesson is creating an intuitive, spatial gesture interface. Imagine looking at your monitor display where a graphical representation of a circular LED ring is overlaid onto your live camera feed. By extending your hand and placing your index finger tip over any of the twelve virtual LED nodes on screen, the system detects your hover position. When you pinch your thumb and index finger together, the program registers a toggle event, turning the corresponding physical LED on or off on the actual hardware ring sitting on your workbench.

To make this work reliably without wild bouncing or rapid flickering, we dive deep into the mechanics of gesture latching and debouncing. When a student first attempts spatial tracking, holding a pinch gesture over a target often causes the state to flip back and forth dozens of times per second. In this lesson, we implement a stateful boolean latch flag that locks down the trigger state the moment a pinch is registered. The algorithm strictly requires you to release the pinch—breaking contact between your thumb and index finger—before it will accept another command. This simple state machine logic is essential for turning noisy computer vision inputs into rock-solid physical control interfaces.

Beyond individual pixel control, we also construct a dynamic UI Mode Selection Panel on the right side of the screen. By hovering over the panel buttons and executing the same pinch gesture, you can cycle through five distinct lighting animation profiles in real time:

Static Mode: Displays fixed, custom-assigned palette colors for each active LED node.
Rainbow Mode: Distributes a static 360-degree color wheel spectrum across all twelve positions using HSV-to-RGB conversion algorithms.
Pulsate Mode: Modulates the physical brightness using a squared sine wave math function to guarantee a true black-floor fade on turned-on pixels.
Blink Mode: Synchronizes an on/off strobe animation derived from live system timestamps.
Running Rainbow Mode: Dynamically rotates the active hue spectrum around the ring by stepping the color offset frame-by-frame.

We handle all of this frame rate processing seamlessly using high-speed SPI bus communications through neopixel_spi and direct frame acquisition via Picamera2. By keeping auto-write disabled until all mathematical animation steps and hand-tracking calculations are completed for a given frame, we push smooth, flicker-free updates out to both the physical hardware and the OpenCV display overlay simultaneously.

Work through the complete Python script provided on this page, load it onto your Pi, and assemble your hardware circuit. Pay close attention to how the spatial math maps coordinates from normalized MediaPipe landmark spaces directly into circular screen coordinates using polar angle formulas. Grab your components, fire up your IDE, and let’s get building!

# ====================================================================
# DISCLAIMER:
# This code is provided as-is for educational and experimental 
# purposes only. The author makes no representations or warranties of 
# any kind concerning the safety, suitability, or accuracy of this 
# code. Use at your own risk. The author assumes no liability for any 
# damages, system failures, security breaches, or network issues 
# resulting from the use or implementation of this script.
# ====================================================================

# Import required computer vision, system, math, and hardware libraries
import cv2
import time
import numpy as np
import colorsys
import board
import neopixel_spi as neopixel
from picamera2 import Picamera2
import mediapipe as mp

# ==============================================================================
# HARDWARE & NEOPIXEL INITIALIZATION
# ==============================================================================

# Initialize SPI communication bus for high-speed LED control
spiBus = board.SPI()

# Total number of physical LEDs in the circular NeoPixel ring
ledCount = 12

# Color order setting required by NeoPixel hardware (Green-Red-Blue)
pixelOrder = neopixel.GRB

# Create the SPI NeoPixel strip object, disabling auto-write for manual updates
ledStrip = neopixel.NeoPixel_SPI(spiBus, ledCount, pixel_order=pixelOrder, auto_write=False)

# Clear any previous LED states by turning all physical pixels off initially
ledStrip.fill(0)
ledStrip.show()

print("✅ NeoPixel initialized!")

# Boolean array tracking the active power state (True = On / False = Off) for each LED
ledStates = [False] * 12

# Array tracking the active RGB color for each physical LED to sync with screen drawing
currentLedRgb = [(0, 0, 0)] * 12

# Static base RGB color mapping assigned to each individual LED index
ledColors = [
    (255, 0, 0),    # LED 0: Red
    (0, 255, 0),    # LED 1: Green
    (0, 0, 255),    # LED 2: Blue
    (0, 255, 255),  # LED 3: Cyan
    (255, 0, 255),  # LED 4: Magenta
    (255, 255, 0),  # LED 5: Yellow
    (255, 100, 0),  # LED 6: Orange
    (128, 0, 255),  # LED 7: Purple
    (0, 255, 128),  # LED 8: Spring Green
    (255, 0, 128),  # LED 9: Pink
    (128, 255, 0),  # LED 10: Chartreuse
    (0, 128, 255)   # LED 11: Light Blue
]

# ==============================================================================
# LIGHTING MODES & STATE TRACKING
# ==============================================================================

# Available animation profiles selectable from the UI panel
modes = ["Static", "Rainbow", "Pulsate", "Blink", "Running Rainbow"]

# Index tracker for the active lighting mode
currentMode = 0

# Math phase tracker used for calculating brightness modulation in Pulsate mode
pulsePhase = 0

# Degree offset tracker used for rotating hue spectrums in Rainbow modes
rainbowOffset = 0

# Degree offset for frame-by-frame 1-degree hue increment
runningHueOffset = 0

# ==============================================================================
# DEBOUNCE & GESTURE LATCHING CONTROL FLAGS
# ==============================================================================

# Stateful latch flag to require full gesture release before accepting new toggles.
# This prevents rapid re-triggering (blinking) when holding a pinch over a node.
isGestureActive = False

# ==============================================================================
# PICAMERA2 & MEDIAPIPE HAND TRACKING SETUP
# ==============================================================================

# Instantiate PiCamera2 object targeting camera device 1
piCam = Picamera2(1)

# Screen resolution settings matching camera stream aspect ratio
frameWidth = 1280
frameHeight = 720

# Configure preview window resolution and color stream formatting
piCam.preview_configuration.main.size = (frameWidth, frameHeight)
piCam.preview_configuration.main.format = "RGB888"
piCam.preview_configuration.controls.FrameRate = 60
piCam.preview_configuration.align()
piCam.configure("preview")
piCam.start()

# Load MediaPipe hand detection module components
mpHands = mp.solutions.hands
handsDetector = mpHands.Hands(
    model_complexity=1, 
    min_detection_confidence=0.7,
    min_tracking_confidence=0.7, 
    max_num_hands=1
)
mpDraw = mp.solutions.drawing_utils

# Visual overlay center coordinate and radius for rendering the circular LED ring
centerX = frameWidth // 2 
centerY = frameHeight // 2
ringRadius = 220
cv2.namedWindow("NeoPixel Ring Controller",cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow("NeoPixel Ring Controller",0,65)
cv2.resizeWindow("NeoPixel Ring Controller",frameWidth,frameHeight)
# ==============================================================================
# MAIN APPLICATION LOOP
# ==============================================================================
while True:
    # Capture raw frame matrix from PiCamera stream
    frameData = piCam.capture_array()
    
    # Mirror preview display horizontally and vertically to match natural orientation
    frameData = cv2.flip(frameData, -1)
    frameData = cv2.flip(frameData, 1)
    
    # Convert image format from BGR to RGB required by MediaPipe
    rgbFrame = cv2.cvtColor(frameData, cv2.COLOR_BGR2RGB)

    # Execute hand keypoint extraction on the converted frame
    trackingResults = handsDetector.process(rgbFrame)

    # Create image copy dedicated for drawing visual UI overlays
    displayFrame = frameData.copy()

    # Capture current UNIX epoch timestamp for time-based animation calculations
    currentTime = time.time()

    # Flag tracking whether a valid action gesture occurred in the current frame
    gestureTriggeredThisFrame = False

    # Process hand gesture logic if hand landmarks are detected
    if trackingResults.multi_hand_landmarks:
        for handLandmarks in trackingResults.multi_hand_landmarks:
            # Draw skeletal connections over the detected hand on display image
            mpDraw.draw_landmarks(displayFrame, handLandmarks, mpHands.HAND_CONNECTIONS)

            # Extract full list of 3D spatial points for all key joints
            landmarksList = handLandmarks.landmark
            indexTipPoint = landmarksList[8]  # Joint landmark index 8 (Index finger tip)
            thumbTipPoint = landmarksList[4]  # Joint landmark index 4 (Thumb tip)

            # Map coordinates from normalized values (0-1) to pixel dimensions
            indexPixelX = int(indexTipPoint.x * frameWidth)
            indexPixelY = int(indexTipPoint.y * frameHeight)
            thumbPixelX = int(thumbTipPoint.x * frameWidth)
            thumbPixelY = int(thumbTipPoint.y * frameHeight)

            # Calculate Euclidean distance between index tip and thumb tip in pixels
            pinchDistance = np.hypot(indexPixelX - thumbPixelX, indexPixelY - thumbPixelY)

            # Check pointing interaction against every LED position along the circle
            for ledIdx in range(12):
                # Angle formula: Starts at +pi/2 (bottom) and rotates clockwise (- direction)
                polarAngle = (np.pi / 2) - ledIdx * (2 * np.pi / 12)
                
                # Derive 2D screen positions matching physical ring layout
                ledPixelX = int(centerX + ringRadius * np.cos(polarAngle))
                ledPixelY = int(centerY + ringRadius * np.sin(polarAngle))
                
                # Compute straight-line Euclidean distance between index finger and target LED
                pointDistance = np.hypot(indexPixelX - ledPixelX, indexPixelY - ledPixelY)

                # Determine if fingertip is hovering within interaction radius of an LED point (Relaxed to 60px)
                if pointDistance < 60:
                    # Highlight active target LED with concentric indicator rings
                    cv2.circle(displayFrame, (ledPixelX, ledPixelY), 38, (0, 255, 255), 6)
                    cv2.circle(displayFrame, (ledPixelX, ledPixelY), 22, (255, 255, 100), -1)

                    # Verify pinch condition (Relaxed to 55px threshold for easier detection)
                    if pinchDistance < 55:
                        gestureTriggeredThisFrame = True
                        
                        # Draw pinch visual feedback ring connecting index and thumb
                        cv2.circle(displayFrame, (indexPixelX, indexPixelY), 15, (0, 255, 0), -1)
                        
                        # Execute state change only if the gesture was not already latched
                        if not isGestureActive:
                            # Invert boolean state for selected LED
                            ledStates[ledIdx] = not ledStates[ledIdx]
                            
                            # Lock gesture activation state until user unpinches
                            isGestureActive = True

            # Process UI mode panel interaction if pointing inside designated top-right screen area
            if indexPixelX > frameWidth - 280:
                if indexPixelY < 420:
                    # Map vertical pixel coordinate directly to corresponding mode list index
                    modeIndexChoice = (indexPixelY - 80) // 70
                    
                    # Verify target index falls within valid bounds of the mode selection list
                    if 0 <= modeIndexChoice < len(modes):
                        # Verify pinch condition to confirm mode selection (Relaxed to 55px)
                        if pinchDistance < 55:
                            gestureTriggeredThisFrame = True
                            
                            # Draw pinch visual feedback ring connecting index and thumb
                            cv2.circle(displayFrame, (indexPixelX, indexPixelY), 15, (0, 255, 0), -1)
                            
                            # Execute mode switch only if the gesture was not already latched
                            if not isGestureActive:
                                currentMode = modeIndexChoice
                                
                                # Lock gesture activation state until user unpinches
                                isGestureActive = True

    # Reset latch flag when hand gesture is unpinched or moved off targets
    if not gestureTriggeredThisFrame:
        isGestureActive = False

    # ==========================================================================
    # DYNAMIC LIGHTING ANIMATION EFFECTS & HARDWARE COLOR UPDATES
    # ==========================================================================
    
    # MODE 0: Static colors assigned directly from preset ledColors palette
    if currentMode == 0:
        for ledIdx in range(12):
            if ledStates[ledIdx]:
                currentLedRgb[ledIdx] = ledColors[ledIdx]
            if not ledStates[ledIdx]:
                currentLedRgb[ledIdx] = (0, 0, 0)

# MODE 1: Fixed rainbow distribution (Red at LED 0, stepping by 30 degrees, static)
    if currentMode == 1:
        for ledIdx in range(12):
            if ledStates[ledIdx]:
                calculatedHue = ledIdx * 30  # 360 degrees / 12 LEDs = 30 degrees per step
                redVal, greenVal, blueVal = [int(channel * 255) for channel in colorsys.hsv_to_rgb(calculatedHue / 360.0, 1, 1)]
                currentLedRgb[ledIdx] = (redVal, greenVal, blueVal)
            if not ledStates[ledIdx]:
                currentLedRgb[ledIdx] = (0, 0, 0)

    # MODE 2: Pulsating brightness modulation (squared sine wave for true black floor fade)
    if currentMode == 2:
        pulsePhase += 0.15
        calculatedBrightness = ((np.sin(pulsePhase) + 1) / 2) ** 2
        for ledIdx in range(12):
            if ledStates[ledIdx]:
                baseRed, baseGreen, baseBlue = ledColors[ledIdx]
                currentLedRgb[ledIdx] = (
                    int(baseRed * calculatedBrightness),
                    int(baseGreen * calculatedBrightness),
                    int(baseBlue * calculatedBrightness)
                )
            if not ledStates[ledIdx]:
                currentLedRgb[ledIdx] = (0, 0, 0)

    # MODE 3: Blinking on/off light strobe animation
    if currentMode == 3:
        blinkToggleState = int(currentTime * 3) % 2
        for ledIdx in range(12):
            if ledStates[ledIdx]:
                if blinkToggleState:
                    currentLedRgb[ledIdx] = ledColors[ledIdx]
                if not blinkToggleState:
                    currentLedRgb[ledIdx] = (0, 0, 0)
            if not ledStates[ledIdx]:
                currentLedRgb[ledIdx] = (0, 0, 0)

# MODE 4: Evenly-spaced rainbow incrementing 1 degree per cycle
    if currentMode == 4:
        for ledIdx in range(12):
            if ledStates[ledIdx]:
                # Start at evenly spaced base angle (30 deg apart) and add 1-degree step offset
                calculatedHue = (ledIdx * 30 + runningHueOffset) % 360

                # Convert normalized float hue (0.0 to 1.0) to RGB
                rFloat, gFloat, bFloat = colorsys.hsv_to_rgb(calculatedHue / 360.0, 1.0, 1.0)
                currentLedRgb[ledIdx] = (
                    int(round(rFloat * 255)),
                    int(round(gFloat * 255)),
                    int(round(bFloat * 255))
                )
            if not ledStates[ledIdx]:
                currentLedRgb[ledIdx] = (0, 0, 0)

        # Increment offset by exactly 1 degree each frame cycle, wrapping 360 back to 0
        runningHueOffset = (runningHueOffset + 2) % 360
    # Push all updated mode colors out to physical NeoPixel hardware strip
    for ledIdx in range(12):
        ledStrip[ledIdx] = currentLedRgb[ledIdx]
    ledStrip.show()

    # ==========================================================================
    # ON-SCREEN GRAPHICAL RENDERING (LED RING MATCHING PHYSICAL HARDWARE)
    # ==========================================================================
    for ledIdx in range(12):
        # Angle formula: Starts at +pi/2 (bottom) and rotates clockwise (- direction)
        polarAngle = (np.pi / 2) - ledIdx * (2 * np.pi / 12)
        targetX = int(centerX + ringRadius * np.cos(polarAngle))
        targetY = int(centerY + ringRadius * np.sin(polarAngle))

        # Render illuminated state representation using exact active hardware color
        if ledStates[ledIdx]:
            # Extract current RGB hardware values
            redVal, greenVal, blueVal = currentLedRgb[ledIdx]
            
            # Convert RGB tuple to BGR tuple format required by OpenCV rendering
            openCvBgrColor = (blueVal, greenVal, redVal)
            
            # Draw illuminated circle matching active physical LED color
            cv2.circle(displayFrame, (targetX, targetY), 24, openCvBgrColor, -1)
            cv2.circle(displayFrame, (targetX, targetY), 31, (255, 255, 255), 4)

        # Render dimmed inactive state representation if turned off
        if not ledStates[ledIdx]:
            cv2.circle(displayFrame, (targetX, targetY), 20, (50, 50, 50), -1)
            cv2.circle(displayFrame, (targetX, targetY), 27, (90, 90, 90), 3)

        # Draw LED index numerical labels inside each visual circle node
        cv2.putText(
            displayFrame, str(ledIdx), (targetX - 9, targetY + 7),
            cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 255, 255), 2
        )

    # ==========================================================================
    # ON-SCREEN GRAPHICAL RENDERING (MODE SELECTION PANEL)
    # ==========================================================================
    panelPositionX = frameWidth - 260
    for modeIdx, modeName in enumerate(modes):
        panelPositionY = 80 + modeIdx * 70
        
        # Default unselected button background color in BGR (Dark Gray)
        highlightColor = (100, 100, 100)
        
        # Override background color with vibrant Blue in BGR format when mode is selected
        if modeIdx == currentMode:
            highlightColor = (255, 100, 0)

        # Draw background menu card rectangle
        cv2.rectangle(displayFrame, (panelPositionX, panelPositionY), (panelPositionX + 240, panelPositionY + 60), highlightColor, -1)
        
        # Draw outer border line around menu card rectangle
        cv2.rectangle(displayFrame, (panelPositionX, panelPositionY), (panelPositionX + 240, panelPositionY + 60), (255, 255, 255), 3)
        
        # Render string name for lighting mode into menu card
        cv2.putText(
            displayFrame, modeName, (panelPositionX + 20, panelPositionY + 40),
            cv2.FONT_HERSHEY_SIMPLEX, 0.75, (255, 255, 255), 2
        )

    # ==========================================================================
    # INSTRUCTION OVERLAYS & WINDOW MANAGEMENT
    # ==========================================================================
    cv2.putText(
        displayFrame, "Point + Pinch = Toggle LED", (30, frameHeight - 40),
        cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 255), 2
    )
    cv2.putText(
        displayFrame, "Point Right Panel + Pinch = Select Mode", (30, frameHeight - 80),
        cv2.FONT_HERSHEY_SIMPLEX, 0.65, (200, 200, 200), 2
    )

    # Render completed frame to OpenCV GUI window
    cv2.imshow("NeoPixel Ring Controller", displayFrame)
    # Read user keypress events (Wait for 1 millisecond)
    activeKey = cv2.waitKey(1) & 0xFF
    
    # Process exit program trigger if 'q' key is pressed
    if activeKey == ord('q'):
        break
        
    # Process reset/clear trigger if 'c' key is pressed
    if activeKey == ord('c'):
        ledStates = [False] * 12
        currentLedRgb = [(0, 0, 0)] * 12
        ledStrip.fill(0)
        ledStrip.show()

# ==============================================================================
# CLEANUP & SHUTDOWN PROCEDURES
# ==============================================================================
print('Program Terminated')

# Close open OpenCV GUI windows
cv2.destroyAllWindows()

# Ensure all physical NeoPixels are safely powered off on exit
ledStrip.fill(0)
ledStrip.show()

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

# ====================================================================

# DISCLAIMER:

# This code is provided as-is for educational and experimental

# purposes only. The author makes no representations or warranties of

# any kind concerning the safety, suitability, or accuracy of this

# code. Use at your own risk. The author assumes no liability for any

# damages, system failures, security breaches, or network issues

# resulting from the use or implementation of this script.

# ====================================================================

# Import required computer vision, system, math, and hardware libraries

import cv2

import time

import numpy as np

import colorsys

import board

import neopixel_spi as neopixel

from picamera2 import Picamera2

import mediapipe as mp

# ==============================================================================

# HARDWARE & NEOPIXEL INITIALIZATION

# ==============================================================================

# Initialize SPI communication bus for high-speed LED control

spiBus = board.SPI()

# Total number of physical LEDs in the circular NeoPixel ring

ledCount = 12

# Color order setting required by NeoPixel hardware (Green-Red-Blue)

pixelOrder = neopixel.GRB

# Create the SPI NeoPixel strip object, disabling auto-write for manual updates

ledStrip = neopixel.NeoPixel_SPI(spiBus, ledCount, pixel_order=pixelOrder, auto_write=False)

# Clear any previous LED states by turning all physical pixels off initially

ledStrip.fill(0)

ledStrip.show()

print("✅ NeoPixel initialized!")

# Boolean array tracking the active power state (True = On / False = Off) for each LED

ledStates = [False] * 12

# Array tracking the active RGB color for each physical LED to sync with screen drawing

currentLedRgb = [(0, 0, 0)] * 12

# Static base RGB color mapping assigned to each individual LED index

ledColors = [

(255, 0, 0), # LED 0: Red

(0, 255, 0), # LED 1: Green

(0, 0, 255), # LED 2: Blue

(0, 255, 255), # LED 3: Cyan

(255, 0, 255), # LED 4: Magenta

(255, 255, 0), # LED 5: Yellow

(255, 100, 0), # LED 6: Orange

(128, 0, 255), # LED 7: Purple

(0, 255, 128), # LED 8: Spring Green

(255, 0, 128), # LED 9: Pink

(128, 255, 0), # LED 10: Chartreuse

(0, 128, 255) # LED 11: Light Blue

]

# ==============================================================================

# LIGHTING MODES & STATE TRACKING

# ==============================================================================

# Available animation profiles selectable from the UI panel

modes = ["Static", "Rainbow", "Pulsate", "Blink", "Running Rainbow"]

# Index tracker for the active lighting mode

currentMode = 0

# Math phase tracker used for calculating brightness modulation in Pulsate mode

pulsePhase = 0

# Degree offset tracker used for rotating hue spectrums in Rainbow modes

rainbowOffset = 0

# Degree offset for frame-by-frame 1-degree hue increment

runningHueOffset = 0

# ==============================================================================

# DEBOUNCE & GESTURE LATCHING CONTROL FLAGS

# ==============================================================================

# Stateful latch flag to require full gesture release before accepting new toggles.

# This prevents rapid re-triggering (blinking) when holding a pinch over a node.

isGestureActive = False

# ==============================================================================

# PICAMERA2 & MEDIAPIPE HAND TRACKING SETUP

# ==============================================================================

# Instantiate PiCamera2 object targeting camera device 1

piCam = Picamera2(1)

# Screen resolution settings matching camera stream aspect ratio

frameWidth = 1280

frameHeight = 720

# Configure preview window resolution and color stream formatting

piCam.preview_configuration.main.size = (frameWidth, frameHeight)

piCam.preview_configuration.main.format = "RGB888"

piCam.preview_configuration.controls.FrameRate = 60

piCam.preview_configuration.align()

piCam.configure("preview")

piCam.start()

# Load MediaPipe hand detection module components

mpHands = mp.solutions.hands

handsDetector = mpHands.Hands(

model_complexity=1,

min_detection_confidence=0.7,

min_tracking_confidence=0.7,

max_num_hands=1

)

mpDraw = mp.solutions.drawing_utils

# Visual overlay center coordinate and radius for rendering the circular LED ring

centerX = frameWidth // 2

centerY = frameHeight // 2

ringRadius = 220

cv2.namedWindow("NeoPixel Ring Controller",cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow("NeoPixel Ring Controller",0,65)

cv2.resizeWindow("NeoPixel Ring Controller",frameWidth,frameHeight)

# ==============================================================================

# MAIN APPLICATION LOOP

# ==============================================================================

while True:

# Capture raw frame matrix from PiCamera stream

frameData = piCam.capture_array()

# Mirror preview display horizontally and vertically to match natural orientation

frameData = cv2.flip(frameData, -1)

frameData = cv2.flip(frameData, 1)

# Convert image format from BGR to RGB required by MediaPipe

rgbFrame = cv2.cvtColor(frameData, cv2.COLOR_BGR2RGB)

# Execute hand keypoint extraction on the converted frame

trackingResults = handsDetector.process(rgbFrame)

# Create image copy dedicated for drawing visual UI overlays

displayFrame = frameData.copy()

# Capture current UNIX epoch timestamp for time-based animation calculations

currentTime = time.time()

# Flag tracking whether a valid action gesture occurred in the current frame

gestureTriggeredThisFrame = False

# Process hand gesture logic if hand landmarks are detected

if trackingResults.multi_hand_landmarks:

for handLandmarks in trackingResults.multi_hand_landmarks:

# Draw skeletal connections over the detected hand on display image

mpDraw.draw_landmarks(displayFrame, handLandmarks, mpHands.HAND_CONNECTIONS)

# Extract full list of 3D spatial points for all key joints

landmarksList = handLandmarks.landmark

indexTipPoint = landmarksList[8] # Joint landmark index 8 (Index finger tip)

thumbTipPoint = landmarksList[4] # Joint landmark index 4 (Thumb tip)

# Map coordinates from normalized values (0-1) to pixel dimensions

indexPixelX = int(indexTipPoint.x * frameWidth)

indexPixelY = int(indexTipPoint.y * frameHeight)

thumbPixelX = int(thumbTipPoint.x * frameWidth)

thumbPixelY = int(thumbTipPoint.y * frameHeight)

# Calculate Euclidean distance between index tip and thumb tip in pixels

pinchDistance = np.hypot(indexPixelX - thumbPixelX, indexPixelY - thumbPixelY)

# Check pointing interaction against every LED position along the circle

for ledIdx in range(12):

# Angle formula: Starts at +pi/2 (bottom) and rotates clockwise (- direction)

polarAngle = (np.pi / 2) - ledIdx * (2 * np.pi / 12)

# Derive 2D screen positions matching physical ring layout

ledPixelX = int(centerX + ringRadius * np.cos(polarAngle))

ledPixelY = int(centerY + ringRadius * np.sin(polarAngle))

# Compute straight-line Euclidean distance between index finger and target LED

pointDistance = np.hypot(indexPixelX - ledPixelX, indexPixelY - ledPixelY)

# Determine if fingertip is hovering within interaction radius of an LED point (Relaxed to 60px)

if pointDistance < 60:

# Highlight active target LED with concentric indicator rings

cv2.circle(displayFrame, (ledPixelX, ledPixelY), 38, (0, 255, 255), 6)

cv2.circle(displayFrame, (ledPixelX, ledPixelY), 22, (255, 255, 100), -1)

# Verify pinch condition (Relaxed to 55px threshold for easier detection)

if pinchDistance < 55:

gestureTriggeredThisFrame = True

# Draw pinch visual feedback ring connecting index and thumb

cv2.circle(displayFrame, (indexPixelX, indexPixelY), 15, (0, 255, 0), -1)

# Execute state change only if the gesture was not already latched

if not isGestureActive:

# Invert boolean state for selected LED

ledStates[ledIdx] = not ledStates[ledIdx]

# Lock gesture activation state until user unpinches

isGestureActive = True

# Process UI mode panel interaction if pointing inside designated top-right screen area

if indexPixelX > frameWidth - 280:

if indexPixelY < 420:

# Map vertical pixel coordinate directly to corresponding mode list index

modeIndexChoice = (indexPixelY - 80) // 70

# Verify target index falls within valid bounds of the mode selection list

if 0 <= modeIndexChoice < len(modes):

# Verify pinch condition to confirm mode selection (Relaxed to 55px)

if pinchDistance < 55:

gestureTriggeredThisFrame = True

# Draw pinch visual feedback ring connecting index and thumb

cv2.circle(displayFrame, (indexPixelX, indexPixelY), 15, (0, 255, 0), -1)

# Execute mode switch only if the gesture was not already latched

if not isGestureActive:

currentMode = modeIndexChoice

# Lock gesture activation state until user unpinches

isGestureActive = True

# Reset latch flag when hand gesture is unpinched or moved off targets

if not gestureTriggeredThisFrame:

isGestureActive = False

# ==========================================================================

# DYNAMIC LIGHTING ANIMATION EFFECTS & HARDWARE COLOR UPDATES

# ==========================================================================

# MODE 0: Static colors assigned directly from preset ledColors palette

if currentMode == 0:

for ledIdx in range(12):

if ledStates[ledIdx]:

currentLedRgb[ledIdx] = ledColors[ledIdx]

if not ledStates[ledIdx]:

currentLedRgb[ledIdx] = (0, 0, 0)

# MODE 1: Fixed rainbow distribution (Red at LED 0, stepping by 30 degrees, static)

if currentMode == 1:

for ledIdx in range(12):

if ledStates[ledIdx]:

calculatedHue = ledIdx * 30 # 360 degrees / 12 LEDs = 30 degrees per step

redVal, greenVal, blueVal = [int(channel * 255) for channel in colorsys.hsv_to_rgb(calculatedHue / 360.0, 1, 1)]

currentLedRgb[ledIdx] = (redVal, greenVal, blueVal)

if not ledStates[ledIdx]:

currentLedRgb[ledIdx] = (0, 0, 0)

# MODE 2: Pulsating brightness modulation (squared sine wave for true black floor fade)

if currentMode == 2:

pulsePhase += 0.15

calculatedBrightness = ((np.sin(pulsePhase) + 1) / 2) ** 2

for ledIdx in range(12):

if ledStates[ledIdx]:

baseRed, baseGreen, baseBlue = ledColors[ledIdx]

currentLedRgb[ledIdx] = (

int(baseRed * calculatedBrightness),

int(baseGreen * calculatedBrightness),

int(baseBlue * calculatedBrightness)

)

if not ledStates[ledIdx]:

currentLedRgb[ledIdx] = (0, 0, 0)

# MODE 3: Blinking on/off light strobe animation

if currentMode == 3:

blinkToggleState = int(currentTime * 3) % 2

for ledIdx in range(12):

if ledStates[ledIdx]:

if blinkToggleState:

currentLedRgb[ledIdx] = ledColors[ledIdx]

if not blinkToggleState:

currentLedRgb[ledIdx] = (0, 0, 0)

if not ledStates[ledIdx]:

currentLedRgb[ledIdx] = (0, 0, 0)

# MODE 4: Evenly-spaced rainbow incrementing 1 degree per cycle

if currentMode == 4:

for ledIdx in range(12):

if ledStates[ledIdx]:

# Start at evenly spaced base angle (30 deg apart) and add 1-degree step offset

calculatedHue = (ledIdx * 30 + runningHueOffset) % 360

# Convert normalized float hue (0.0 to 1.0) to RGB

rFloat, gFloat, bFloat = colorsys.hsv_to_rgb(calculatedHue / 360.0, 1.0, 1.0)

currentLedRgb[ledIdx] = (

int(round(rFloat * 255)),

int(round(gFloat * 255)),

int(round(bFloat * 255))

)

if not ledStates[ledIdx]:

currentLedRgb[ledIdx] = (0, 0, 0)

# Increment offset by exactly 1 degree each frame cycle, wrapping 360 back to 0

runningHueOffset = (runningHueOffset + 2) % 360

# Push all updated mode colors out to physical NeoPixel hardware strip

for ledIdx in range(12):

ledStrip[ledIdx] = currentLedRgb[ledIdx]

ledStrip.show()

# ==========================================================================

# ON-SCREEN GRAPHICAL RENDERING (LED RING MATCHING PHYSICAL HARDWARE)

# ==========================================================================

for ledIdx in range(12):

# Angle formula: Starts at +pi/2 (bottom) and rotates clockwise (- direction)

polarAngle = (np.pi / 2) - ledIdx * (2 * np.pi / 12)

targetX = int(centerX + ringRadius * np.cos(polarAngle))

targetY = int(centerY + ringRadius * np.sin(polarAngle))

# Render illuminated state representation using exact active hardware color

if ledStates[ledIdx]:

# Extract current RGB hardware values

redVal, greenVal, blueVal = currentLedRgb[ledIdx]

# Convert RGB tuple to BGR tuple format required by OpenCV rendering

openCvBgrColor = (blueVal, greenVal, redVal)

# Draw illuminated circle matching active physical LED color

cv2.circle(displayFrame, (targetX, targetY), 24, openCvBgrColor, -1)

cv2.circle(displayFrame, (targetX, targetY), 31, (255, 255, 255), 4)

# Render dimmed inactive state representation if turned off

if not ledStates[ledIdx]:

cv2.circle(displayFrame, (targetX, targetY), 20, (50, 50, 50), -1)

cv2.circle(displayFrame, (targetX, targetY), 27, (90, 90, 90), 3)

# Draw LED index numerical labels inside each visual circle node

cv2.putText(

displayFrame, str(ledIdx), (targetX - 9, targetY + 7),

cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 255, 255), 2

)

# ==========================================================================

# ON-SCREEN GRAPHICAL RENDERING (MODE SELECTION PANEL)

# ==========================================================================

panelPositionX = frameWidth - 260

for modeIdx, modeName in enumerate(modes):

panelPositionY = 80 + modeIdx * 70

# Default unselected button background color in BGR (Dark Gray)

highlightColor = (100, 100, 100)

# Override background color with vibrant Blue in BGR format when mode is selected

if modeIdx == currentMode:

highlightColor = (255, 100, 0)

# Draw background menu card rectangle

cv2.rectangle(displayFrame, (panelPositionX, panelPositionY), (panelPositionX + 240, panelPositionY + 60), highlightColor, -1)

# Draw outer border line around menu card rectangle

cv2.rectangle(displayFrame, (panelPositionX, panelPositionY), (panelPositionX + 240, panelPositionY + 60), (255, 255, 255), 3)

# Render string name for lighting mode into menu card

cv2.putText(

displayFrame, modeName, (panelPositionX + 20, panelPositionY + 40),

cv2.FONT_HERSHEY_SIMPLEX, 0.75, (255, 255, 255), 2

)

# ==========================================================================

# INSTRUCTION OVERLAYS & WINDOW MANAGEMENT

# ==========================================================================

cv2.putText(

displayFrame, "Point + Pinch = Toggle LED", (30, frameHeight - 40),

cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 255), 2

)

cv2.putText(

displayFrame, "Point Right Panel + Pinch = Select Mode", (30, frameHeight - 80),

cv2.FONT_HERSHEY_SIMPLEX, 0.65, (200, 200, 200), 2

)

# Render completed frame to OpenCV GUI window

cv2.imshow("NeoPixel Ring Controller", displayFrame)

# Read user keypress events (Wait for 1 millisecond)

activeKey = cv2.waitKey(1) & 0xFF

# Process exit program trigger if 'q' key is pressed

if activeKey == ord('q'):

break

# Process reset/clear trigger if 'c' key is pressed

if activeKey == ord('c'):

ledStates = [False] * 12

currentLedRgb = [(0, 0, 0)] * 12

ledStrip.fill(0)

ledStrip.show()

# ==============================================================================

# CLEANUP & SHUTDOWN PROCEDURES

# ==============================================================================

print('Program Terminated')

# Close open OpenCV GUI windows

cv2.destroyAllWindows()

# Ensure all physical NeoPixels are safely powered off on exit

ledStrip.fill(0)

ledStrip.show()

Technology Tutorials

AI on the Edge LESSON 54: Custom Hand Gesture Recognition and Dynamic NeoPixel Control

What We Are Building Today

Hardware Setup & Wiring (SPI Interface)

Wiring Diagram Guide

⌨️ Interactive Keyboard Controls

Complete Python Source Code

Homework Assignment

AI on the Edge LESSON 53: Gesture Recognition on the Pi 5 with Training File

AI on the Edge LESSON 52: Gesture Recognition on the Raspberry Pi 5

Hand Recognition on the Raspberry Pi 5 With openCV and Mediapipe

AI on the Edge LESSON 51: Gesture Control of NeoPixel Ring

Making The World a Better Place One High Tech Project at a Time. Enjoy!