Research Of Key Frames Extrations Algorithms Based On Color Video Monitoring In Complex Background Of Hazard

Along with the rapid development of network technology, multimedia technology and video retrieval technology, the use of multimedia information data is also becoming more and more important in social life, especially video information data processing. It is very important of realistic significance that keeping the video monitoring information of major hazard integrity and real-time, and there is great important theoretic and application value to the development of video monitoring field.Based on the real-time monitoring video information to the hazard, the data collected is usually very huge, so the work on transferring and processing these huge data becomes more difficult. How to solve this problem is the current research hot spot.This paper focuses on the video segment and key frame extraction techniques. Here one method of extracting key frames is improved based on the characteristics of frame sequences in MPEG compression streaming video, mainly studied the characteristics of I frame, P frames and B frame in GOP.To get intuitive surveillance video streams of targeted and high-quality information, reduce the pressure on network traffic and lessen disk storage space, a real-time video monitoring system is designed. A key-frame extraction algorithm is used for MPEG compressed video stream processing, and RTP protocol is used for video transmission. Then, some problems, such as super-large video data, wide bandwidth, and video information redundancy, are successfully resolved. System test results show that the system can be well used for making the video data accepted by the server more focused, has high real-time video transmission, good quality of video restoration, high viability and robustness, and has good value in application.