关于#opencv和yolov5#的问题：将从视频中获取到的信息存在数据库和表格中，存入重复的信息(语言-python)

import cv2

cap = cv2.VideoCapture('path/to/video.mp4')

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

    # 在这里进行YOLOv5物体检测的逻辑
    # ...

cap.release()
cv2.destroyAllWindows()

# 在帧上进行YOLOv5物体检测的逻辑
# 返回的是检测到的物体信息
def perform_yolo_detection(frame):
    # ...
    return objects_detected

import sqlite3

# 建立数据库连接
conn = sqlite3.connect('path/to/database.db')
cursor = conn.cursor()

# 建立表格 (如果尚不存在)
cursor.execute('''CREATE TABLE IF NOT EXISTS detections
                  (frame_number INT, object_name TEXT, x INT, y INT, width INT, height INT)''')

# 存储第一帧的物体信息
frame_number = 0
objects_detected = perform_yolo_detection(frame)

for obj in objects_detected:
    # 将物体信息存入数据库
    cursor.execute('INSERT INTO detections VALUES (?, ?, ?, ?, ?, ?)',
                   (frame_number, obj['name'], obj['x'], obj['y'], obj['width'], obj['height']))
    # 将物体信息存入表格 (例如CSV)
    # ...

conn.commit()

# 关闭连接
conn.close()

import numpy as np

# 存储前一帧
previous_frame = None

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

    # 计算帧差
    if previous_frame is not None:
        frame_diff = cv2.absdiff(previous_frame, frame)
        frame_diff_gray = cv2.cvtColor(frame_diff, cv2.COLOR_BGR2GRAY)
        ret, frame_diff_thresh = cv2.threshold(frame_diff_gray, 30, 255, cv2.THRESH_BINARY)

        # 如果帧差大于阈值，进行YOLOv5物体检测
        if np.sum(frame_diff_thresh) > threshold:
            objects_detected = perform_yolo_detection(frame)

            # 存储物体信息
            frame_number += 1
            for obj in objects_detected:
                # 将物体信息存入数据库
                cursor.execute('INSERT INTO detections VALUES (?, ?, ?, ?, ?, ?)',
                               (frame_number, obj['name'], obj['x'], obj['y'], obj['width'], obj['height']))
                # 将物体信息存入表格 (例如CSV)
                # ...

            conn.commit()

    previous_frame = frame

cap.release()
cv2.destroyAllWindows()