Python-opencv-将指定路径全部视频切帧处理_opencv-python 视频拆帧-优快云博客

本文链接：https://blog.youkuaiyun.com/weixin_50978072/article/details/129125973

该代码段读取指定文件夹中的视频文件，使用OpenCV库打开并处理每一帧。每30帧保存一帧图像到labels文件夹，并通过logging模块记录处理进度。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >


import cv2
import logging
import os

# log information settings
logging.basicConfig(level=logging.INFO,
                    format='%(asctime)s - %(levelname)s: %(message)s')


def save_image(num, image):
    """Save the data.

    Args:
        num: serial number
        image: image resource

    Returns:
        None
    """
    # 指定输出路径
    image_path = './labels/{}.jpg'.format(str(num))
    cv2.imwrite(image_path, image)

count = 0  # count the number of pictures

# 输入视频路径
folder_path = f'./video'
# 遍历文件夹下的所有文件
for root, dirs, files in os.walk(folder_path):
    for file in files:
        # 打印文件路径
        file_path = os.path.join(root, file)


        vc = cv2.VideoCapture(file_path)  # import video files
        print(file_path)
        # determine whether to open normally
        if vc.isOpened():
            ret, frame = vc.read()
        else:
            ret = False


        frame_interval = 30  # video frame count interval frequency
        frame_interval_count = 0

        # loop read video frame
        while ret:
            ret, frame = vc.read()
            # store operation every time f frame
            if frame_interval_count % frame_interval == 0:
                save_image(count, frame)
                logging.info("num：" + str(count) + ", frame: " +
                             str(frame_interval_count))
                count += 1
            frame_interval_count += 1
            cv2.waitKey(5)

        vc.release()