使用OpenCv，对文件夹内的视频文件进行分帧操作，在VSCODE中可以直接运行（更改为自己的文件夹地址），在Pycharm中改为自己的相对地址或者绝对地址，去掉r
import cv2
import os
import time
import datetime
import random
"""
Read video and convert it into pictures
读取视频并将视频转换为图片
注意请勿在视频文件夹中放其他文件
"""
def save_img():
# video_path = r'E:/VideoDir/'
# videos = os.listdir(video_path)
videos = load_video(r'E:/VideoDir/')
for video_name in videos:
vc = cv2.VideoCapture(video_name) # 读入视频文件
c = 1
if vc.isOpened(): # 判断是否正常打开
rval, frame = vc.read()
else:
rval = False
timeF = 15 # 视频帧计数间隔频率
while rval: # 循环读取视频帧
rval, frame = vc.read()
if (c % timeF == 0): # 每隔timeF帧进行存储操作
if frame is not None:
cv2.imwrite("E:/EVideoDir2/" + str(random.randint(1, 100000)) + '_' + str(c) +
'.jpg', frame)
c = c + 1
cv2.waitKey(1)
vc.release()
def load_video(video_path):
input_path_extension = video_path.split('.')[-1]
if input_path_extension in ['mp4']:
return [video_path]
elif input_path_extension == "txt":
with open(video_path, "r") as f:
return f.read().splitlines()
else:
return glob.glob(os.path.join(video_path, "*.mp4"))
if __name__ == '__main__':
save_img()
