通过cv库智能切片 把不同的分镜切出来 自媒体抖音快手混剪

用 手机自动化脚本，从自媒体上获取视频，一个商品对应几百个视频，我们把这几百个视频下载下来，进行分镜

视频切片，从自媒体上下载视频，通过cv库用直方图识别每个镜头进行切片。

下载多个图片进行视频的伪原创，加上智能Ai文字

混剪完之后再通过自动化技术上传到各大平台，这其中真正做到一个人管上千个账号

电商创业团队需要技术的对接

import cv2
import numpy as np


def calculate_histogram(frame):
    gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    hist = cv2.calcHist([gray_frame], [0], None, [256], [0, 256])
    return hist


def histogram_difference(hist1, hist2):
    return cv2.compareHist(hist1, hist2, cv2.HISTCMP_CORREL)

	def main(video_path, output_folder, max_duration=3):

    cap = cv2.VideoCapture(video_path)

    if not cap.isOpened():
        print("无法打开视频文件")
        return

    fps = cap.get(cv2.CAP_PROP_FPS)
    max_frames = int(max_duration * fps)

    ret, prev_frame = cap.read()
    if not ret:
        print("无法读取视频")
        return

    prev_hist = calculate_histogram(prev_frame)
    frame_count = 0
    scene_count = 0
    output_video = None

    while True:
        ret, frame = cap.read()
        if not ret:
            break

        frame_count += 1
        curr_hist = calculate_histogram(frame)
        similarity = histogram_difference(prev_hist, curr_hist)

        if similarity < 0.9:
            if output_video is not None:
                output_video.release()  # 释放之前的输出视频

            output_path = f"{output_folder}/scene_{scene_count}.mp4"
            output_video = cv2.VideoWriter(output_path, cv2.VideoWriter_fourcc(*'mp4v'), fps,
                                           (int(cap.get(3)), int(cap.get(4))))
            scene_count += 1
            frame_count = 0  # 重置帧计数

        # 写入当前帧
        if output_video is not None:
            output_video.write(frame)

        if frame_count >= max_frames and output_video is not None:
            output_video.release()
            output_video = None

        prev_hist = curr_hist

    # 释放资源
    cap.release()
    if output_video is not None:
        output_video.release()


if __name__ == "__main__":
    main("linshi.mp4", "output_folder")