In order to achieve more efficient and accurate remote monitoring,a design method for an embedded intelligent remote video monitoring system is proposed,and facial recognition and license plate recognition algorithms are improved.The hardware of the system includes embedded development boards,cameras,wireless network cards,storage cards,as well as integrated network ports and interfaces on the development board.The test results show that the video transmission delay is within the range of 0-100 ms,and the video transmission stability is within the range of 90％-100％.The system can stably transmit video data,ensuring the accuracy and reliability of monitoring.The accuracy of the proposed deep separable center difference convolutional network algorithm in facial and license plate recognition is 98.9％and 99.2％,respectively,which are superior to convolutional neural and center difference convolutional network algorithms.The system function score of the intelligent remote video monitoring system is 92 points,which is 33 points higher than the original system.The results show that the system has good real-time monitoring performance and video analysis ability,which can improve the efficiency and accuracy of monitoring and provide a reliable solution for monitoring applications.
关键词
嵌入式/树莓派4B/远程/视频监控/高清摄像头/视频分析
Key words
embedded/raspberry Pi 4B/remote/video surveillance/high definition camera/video analysis
维普
万方数据