Python Opencv实践 - 手势音量控制

本文主要是介绍Python Opencv实践 - 手势音量控制，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

本文基于前面的手部跟踪功能做一个手势音量控制功能，代码用到了前面手部跟踪封装的HandDetector.这篇文章在这里：

Python Opencv实践 - 手部跟踪-CSDN博客文章浏览阅读626次，点赞11次，收藏7次。使用mediapipe库做手部的实时跟踪，关于mediapipe的介绍,请自行百度。https://blog.csdn.net/vivo01/article/details/135071340?spm=1001.2014.3001.5502

使用了pycaw来做音量控制，pacaw的安装直接使用pip install pycaw即可。

代码如下：

import cv2 as cv
import math
import mediapipe as mp
import time
from ctypes import cast,POINTER
from comtypes import CLSCTX_ALL
#使用pycaw来控制音量，pip install pycaw
from pycaw.pycaw import AudioUtilities,IAudioEndpointVolumeclass HandDetector():def __init__(self, mode=False,maxNumHands=2,modelComplexity=1,minDetectionConfidence=0.5,minTrackingConfidence=0.5):self.mode = modeself.maxNumHands = maxNumHandsself.modelComplexity = modelComplexityself.minDetectionConfidence = minDetectionConfidenceself.minTrackingConfidence = minTrackingConfidence#创建mediapipe的solutions.hands对象self.mpHands = mp.solutions.handsself.handsDetector = self.mpHands.Hands(self.mode, self.maxNumHands, self.modelComplexity, self.minDetectionConfidence, self.minTrackingConfidence)#创建mediapipe的绘画工具self.mpDrawUtils = mp.solutions.drawing_utilsdef findHands(self, img, drawOnImage=True):#mediapipe手部检测器需要输入图像格式为RGB#cv默认的格式是BGR，需要转换imgRGB = cv.cvtColor(img, cv.COLOR_BGR2RGB)#调用手部检测器的process方法进行检测self.results = self.handsDetector.process(imgRGB)#print(results.multi_hand_landmarks)#如果multi_hand_landmarks有值表示检测到了手if self.results.multi_hand_landmarks:#遍历每一只手的landmarksfor handLandmarks in self.results.multi_hand_landmarks:if drawOnImage:self.mpDrawUtils.draw_landmarks(img, handLandmarks, self.mpHands.HAND_CONNECTIONS)return img;#从结果中查询某只手的landmark listdef findHandPositions(self, img, handID=0, drawOnImage=True):landmarkList = []if self.results.multi_hand_landmarks:handLandmarks = self.results.multi_hand_landmarks[handID]for id,landmark in enumerate(handLandmarks.landmark):#处理每一个landmark,将landmark里的X,Y（比例）转换为帧数据的XY坐标h,w,c = img.shapecenterX,centerY = int(landmark.x * w), int(landmark.y * h)landmarkList.append([id, centerX, centerY])if (drawOnImage):#将landmark绘制成圆cv.circle(img, (centerX,centerY), 8, (0,255,0))return landmarkListdef DisplayFPS(img, preTime):curTime = time.time()if (curTime - preTime == 0):return curTime;fps = 1 / (curTime - preTime)cv.putText(img, "FPS:" + str(int(fps)), (10,70), cv.FONT_HERSHEY_PLAIN,3, (0,255,0), 3)return curTimedef AudioEndpointGet():devices = AudioUtilities.GetSpeakers()interface = devices.Activate(IAudioEndpointVolume._iid_, CLSCTX_ALL, None)volume = cast(interface, POINTER(IAudioEndpointVolume))range = volume.GetVolumeRange()return volume,rangedef AudioVolumeLevelSet(volume, range, value):if volume:if (value < range[0]) or (value > range[1]):returnvolume.SetMasterVolumeLevel(value, None)def main():video = cv.VideoCapture('../../SampleVideos/handVolumeControl.mp4')#FPS显示preTime = 0handDetector = HandDetector(minDetectionConfidence=0.7)volume,volumeRange = AudioEndpointGet()print(volumeRange)#AudioVolumeLevelSet(volume, volumeRange, volumeRange[0])minFingerDistance = 1000maxFingerDistance = 0while True:ret,frame = video.read()if ret == False:break;frame = handDetector.findHands(frame)hand0Landmarks = handDetector.findHandPositions(frame)if (len(hand0Landmarks) != 0):#print(hand0Landmarks[4], hand0Landmarks[8])#取出大拇指(4)和食指(8)的指尖的点对应的坐标thumbX,thumbY = hand0Landmarks[4][1], hand0Landmarks[4][2]indexFingerX,indexFingerY = hand0Landmarks[8][1],hand0Landmarks[8][2]#计算两个指尖的点指尖的中点cx,cy = (thumbX + indexFingerX) / 2, (thumbY + indexFingerY) / 2#用实心圆突出显示出这两个个点cv.circle(frame, (thumbX,thumbY), 18, (90,220,180), cv.FILLED)cv.circle(frame, (indexFingerX,indexFingerY), 18, (0,120,255), cv.FILLED)#绘制两个点形成的直线cv.line(frame, (thumbX,thumbY), (indexFingerX,indexFingerY), (255,60,60), 3)#计算食指和拇指指尖的距离distance = math.hypot(thumbX - indexFingerX, thumbY - indexFingerY)#测试两指指尖最小和最大距离，改进方案可以是用摄像头做实时校准后再进行控制#本案例中直接获取视频里的最小和最大距离直接用作判断（我拍的视频里范围是30 - 425之间）if distance < minFingerDistance:minFingerDistance = distanceif distance > maxFingerDistance:maxFingerDistance = distance#print(distance)if distance < 40:#两个指尖的中点显示为绿色，音量设置为最小值cv.circle(frame, (int(cx),int(cy)), 18, (0,255,0), cv.FILLED)AudioVolumeLevelSet(volume, volumeRange, volumeRange[0])else:cv.circle(frame, (int(cx),int(cy)), 18, (0,0,255), cv.FILLED)#这里为了方便直接使用425（本视频最大值）做比例换算#我本机的volumeRange是-63.5 到 0， 步长0.5value = volumeRange[0] * (1 - (distance / 425))print(value)AudioVolumeLevelSet(volume, volumeRange, value)preTime = DisplayFPS(frame, preTime)cv.imshow('Real Time Hand Detection', frame)if cv.waitKey(30) & 0xFF == ord('q'):break;print("Min & Max distance between thumb and index finger tips: ", minFingerDistance, maxFingerDistance)video.release()cv.destroyAllWindows()if __name__ == "__main__":main()

效果可以参考我的B站视频：

Python Opencv练手-手势音量控制_哔哩哔哩_bilibili基于mediapipe手部检测实现一个手势音量控制功能源码参考我的CSDN：https://blog.csdn.net/vivo01/article/details/135118979?spm=1001.2014.3001.5502, 视频播放量 1、弹幕量 0、点赞数 0、投硬币枚数 0、收藏人数 0、转发人数 0, 视频作者 vivo119, 作者简介一个喜欢小狗子的码农，业余爱好游戏开发，相关视频：小乖最喜欢吃面条，小乖(白）芝麻(黑）的日常冲突，这只胖狗想要跳上沙发，可是胖了点，Python Opencv - mediapipe做手部跟踪识别，为什么小狗看镜头就尴尬，突然爱吃番茄的狗子，旋转的米糯狗子，有手动旋转和自动旋转两种模式，好好上课，小狗的无糖藕粉初体验，米糯狗子洗澡记，全程都是乖乖狗https://www.bilibili.com/video/BV1Ej411H79q/?vd_source=474bff49614e62744eb84e9f8340d91a