python 检测视频区域是否有变化
#!/usr/bin/env python # coding: utf-8 # @author: sSWans # @file: main.py # @time: 2018/1/11 15:54 import os import random from _datetime import datetime import cv2 path = 'd:\\C理论\\ti' # 遍历目录下的视频文件 def get_files(fpath): files_list = [] for i in os.listdir(fpath): files_list.append(os.path.join(fpath, i)) return files_list # 视频处理 def process(file, fname): # camera = cv2.VideoCapture(0) # 参数0表示第一个摄像头 camera = cv2.VideoCapture(file) # 参数设置,监测矩形区域 rectangleX = 750 # 矩形最左点x坐标 rectangleXCols = 340 # 矩形x轴上的长度 rectangleY = 320 # 矩形最上点y坐标 rectangleYCols = 30 # 矩形y轴上的长度 KeyFrame = 17 # 取关键帧的间隔数,根据视频的帧率设置,我的视频是16FPS counter = 1 # 取帧计数器 pre_frame = None # 总是取视频流前一帧做为背景相对下一帧进行比较 # 判断视频是否打开 if not camera.isOpened(): print('视频文件打开失败!') width = int(camera.get(cv2.CAP_PROP_FRAME_WIDTH)) height = int(camera.get(cv2.CAP_PROP_FRAME_HEIGHT)) print('视频尺寸(高,宽):', height, width) if rectangleXCols == 0: rectangleXCols = width - rectangleX if rectangleYCols == 0: rectangleYCols = height - rectangleY start_time = datetime.now() print('{} 开始处理文件: {}'.format(start_time.strftime('%H:%M:%S'), fname)) while True: grabbed, frame_lwpCV = camera.read() # 读取视频流 if grabbed: if counter % KeyFrame == 0: # if not grabbed: # print('{} 完成处理文件: {} 。。。 '.format(datetime.now().strftime('%H:%M:%S'),fname)) # break gray_lwpCV = cv2.cvtColor(frame_lwpCV, cv2.COLOR_BGR2GRAY) # 转灰度图 gray_lwpCV = gray_lwpCV[rectangleY:rectangleY + rectangleYCols, rectangleX:rectangleX + rectangleXCols] lwpCV_box = cv2.rectangle(frame_lwpCV, (rectangleX, rectangleY), (rectangleX + rectangleXCols, rectangleY + rectangleYCols), (0, 255, 0), 2) # 用绿色矩形框显示监测区域 # cv2.imshow('lwpCVWindow', frame_lwpCV) # 显示视频播放窗口,开启消耗时间大概是3倍 gray_lwpCV = cv2.GaussianBlur(gray_lwpCV, (21, 21), 0) if pre_frame is None: pre_frame = gray_lwpCV else: img_delta = cv2.absdiff(pre_frame, gray_lwpCV) thresh = cv2.threshold(img_delta, 25, 255, cv2.THRESH_BINARY)[1] thresh = cv2.dilate(thresh, None, iterations=2) #image, contours, hierarchy = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE) contours, hierarchy = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE) for x in contours: if cv2.contourArea(x) < 1000: # 设置敏感度 continue else: cv2.imwrite( 'image/' + fname + '_' + datetime.now().strftime('%H%M%S') + '_' + str( random.randrange(0, 9999)) + '.jpg', frame_lwpCV) # print("监测到移动物体。。。 ", datetime.now().strftime('%H:%M:%S')) break pre_frame = gray_lwpCV counter += 1 key = cv2.waitKey(1) & 0xFF if key == ord('q'): break else: end_time = datetime.now() print('{} 完成处理文件: {} 耗时:{}'.format(end_time.strftime('%H:%M:%S'), fname, end_time - start_time)) break camera.release() # cv2.destroyAllWindows() # 与上面的imshow对应 for file in get_files(path): fname = file.split('\\')[-1].replace('.mp4', '') process(file, fname)