opencv数字手势识别

数字手势识别是计算机视觉中的一个重要应用，而OpenCV正是一个强大的计算机视觉库，可以用于数字手势识别。下面是一个简单的数字手势识别的例子：

import cv2
import numpy as np

# 定义一些常量
cap_region_x_begin = 0.5  # 起始x坐标
cap_region_y_end = 0.8  # 结束y坐标
threshold = 60  # 二值化阈值
blurValue = 41  # 高斯模糊参数
bgSubThreshold = 50
learningRate = 0

# 定义ROI区域
def nothing(x):
    pass

cv2.namedWindow('trackbar')
cv2.createTrackbar('x', 'trackbar', 0, 100, nothing)
cv2.createTrackbar('y', 'trackbar', 0, 100, nothing)

# 读取摄像头
camera = cv2.VideoCapture(0)
camera.set(10, 200)

# 背景减除器
fgbg = cv2.createBackgroundSubtractorMOG2(detectShadows=True)

while camera.isOpened():
    ret, frame = camera.read()
    frame = cv2.bilateralFilter(frame, 5, 50, 100)  # 双边滤波
    frame = cv2.flip(frame, 1)  # 翻转图像

    # 获取ROI区域
    x = cv2.getTrackbarPos('x', 'trackbar')
    y = cv2.getTrackbarPos('y', 'trackbar')
    roi = frame[int(cap_region_y_end * frame.shape[0]):frame.shape[0], x:frame.shape[1]]

    # 背景减除
    fgmask = fgbg.apply(roi, learningRate=learningRate)
    kernel = np.ones((3, 3), np.uint8)
    fgmask = cv2.erode(fgmask, kernel, iterations=1)
    res = cv2.bitwise_and(roi, roi, mask=fgmask)

    # 灰度化和二值化
    gray = cv2.cvtColor(res, cv2.COLOR_BGR2GRAY)
    blur = cv2.GaussianBlur(gray, (blurValue, blurValue), 0)
    ret, thresh = cv2.threshold(blur, threshold, 255, cv2.THRESH_BINARY)

    # 查找轮廓
    contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

    # 找到最大轮廓
    max_area = 0
    ci = 0
    if contours:
        for i in range(len(contours)):
            cnt = contours[i]
            area = cv2.contourArea(cnt)
            if area > max_area:
                max_area = area
                ci = i
        cnt = contours[ci]

        # 手势识别
        hull = cv2.convexHull(cnt)
        drawing = np.zeros(roi.shape, np.uint8)
        cv2.drawContours(drawing, [cnt], 0, (0, 255, 0), 2)
        cv2.drawContours(drawing, [hull], 0, (0, 0, 255), 3)

        # 计算凸包和凸缺陷
        hull = cv2.convexHull(cnt, returnPoints=False)
        defects = cv2.convexityDefects(cnt, hull)

        # 绘制凸缺陷
        count_defects = 0
        if defects is not None:
            for i in range(defects.shape[0]):
                s, e, f, d = defects[i, 0]
                start = tuple(cnt[s][0])
                end = tuple(cnt[e][0])
                far = tuple(cnt[f][0])
                a = np.sqrt((end[0] - start[0]) ** 2 + (end[1] - start[1]) ** 2)
                b = np.sqrt((far[0] - start[0]) ** 2 + (far[1] - start[1]) ** 2)
                c = np.sqrt((end[0] - far[0]) ** 2 + (end[1] - far[1]) ** 2)
                angle = np.arccos((b ** 2 + c ** 2 - a ** 2) / (2 * b * c)) * 180 / np.pi
                if angle <= 90:
                    count_defects += 1
                    cv2.circle(drawing, far, 8, [211, 84, 0], -1)

        # 显示结果
        cv2.imshow('output', drawing)
        cv2.imshow('input', roi)

    # 按下q键退出
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

camera.release()
cv2.destroyAllWindows()