Hand Gesture Recognition with OpenCV: Video Capture, Skin Detection, and Contour Processing in Python

Recently a beginner‑friendly hand‑gesture processing project was developed using OpenCV, basic Python syntax, and fundamental image‑processing concepts.

Getting Video (Camera)

The video source is obtained either from a video file or directly from a webcam.

<code>cap = cv2.VideoCapture("C:/Users/lenovo/Videos/1.mp4")  # read file
# cap = cv2.VideoCapture(0)  # read webcam
while True:
    ret, frame = cap.read()
    key = cv2.waitKey(50) & 0xFF
    if key == ord('q'):
        break
cap.release()
cv2.destroyAllWindows()</code>

Skin Detection

The algorithm converts the image from BGR to YCrCb color space to reduce the influence of luminance, then applies Gaussian blur and Otsu thresholding to isolate skin pixels.

<code>def A(img):
    YCrCb = cv2.cvtColor(img, cv2.COLOR_BGR2YCR_CB)  # convert to YCrCb
    (y, cr, cb) = cv2.split(YCrCb)                     # split channels
    cr1 = cv2.GaussianBlur(cr, (5, 5), 0)
    _, skin = cv2.threshold(cr1, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)  # Otsu
    res = cv2.bitwise_and(img, img, mask=skin)
    return res</code>

Contour Processing

Contours are extracted with cv2.findContours , sorted by area, and the largest contour (assumed to be the hand) is drawn on a white canvas.

<code>def B(img):
    # binaryimg = cv2.Canny(Laplacian, 50, 200)  # optional binarization
    h = cv2.findContours(img, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)  # find contours
    contour = h[0]
    contour = sorted(contour, key=cv2.contourArea, reverse=True)  # sort by area
    bg = np.ones(dst.shape, np.uint8) * 255  # white canvas
    ret = cv2.drawContours(bg, contour[0], -1, (0, 0, 0), 3)  # draw largest contour
    return ret</code>

Full Script

<code>"""Read frames from video and save as images"""
import cv2
import numpy as np

cap = cv2.VideoCapture("C:/Users/lenovo/Videos/1.mp4")  # read file
# cap = cv2.VideoCapture(0)  # read webcam

def A(img):
    YCrCb = cv2.cvtColor(img, cv2.COLOR_BGR2YCR_CB)
    (y, cr, cb) = cv2.split(YCrCb)
    cr1 = cv2.GaussianBlur(cr, (5, 5), 0)
    _, skin = cv2.threshold(cr1, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    res = cv2.bitwise_and(img, img, mask=skin)
    return res

def B(img):
    # binaryimg = cv2.Canny(Laplacian, 50, 200)
    h = cv2.findContours(img, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
    contour = h[0]
    contour = sorted(contour, key=cv2.contourArea, reverse=True)
    bg = np.ones(dst.shape, np.uint8) * 255
    ret = cv2.drawContours(bg, contour[0], -1, (0, 0, 0), 3)
    return ret

while True:
    ret, frame = cap.read()
    src = cv2.resize(frame, (400, 350), interpolation=cv2.INTER_CUBIC)
    cv2.rectangle(src, (90, 60), (300, 300), (0, 255, 0))
    roi = src[60:300, 90:300]  # hand region
    res = A(roi)               # skin detection
    cv2.imshow("0", roi)
    gray = cv2.cvtColor(res, cv2.COLOR_BGR2GRAY)
    dst = cv2.Laplacian(gray, cv2.CV_16S, ksize=3)
    Laplacian = cv2.convertScaleAbs(dst)
    contour = B(Laplacian)    # contour processing
    cv2.imshow("2", contour)
    key = cv2.waitKey(50) & 0xFF
    if key == ord('q'):
        break
cap.release()
cv2.destroyAllWindows()</code>

The resulting application captures video frames, isolates the hand region using color‑space conversion and thresholding, extracts the dominant contour, and displays both the original ROI and the processed contour image.