Video abstraction method and system based on moving target detection
A moving target and video summarization technology, applied in image data processing, special data processing applications, instruments, etc., can solve the problems of consuming a lot of manpower, material resources and time, storing a large amount of data, missing the opportunity to solve the case, etc., to ensure a compact space performance, efficient compression, target tracking synthesis method simple and effective effect
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Publication Date
- 2013-02-13
Smart Images
Figure 1 Figure 2 Figure 3
Abstract
Description
technical field
[0001] The invention relates to the field of surveillance video summarization and enrichment, in particular to a video summarization method and system based on moving target detection. Background technique
[0002] At present, with the development of economy and information technology, people's demand for security protection is getting higher and higher. As an effective means in the field of security protection, video surveillance has a wider and wider range of applications, and the application requirements are also increasing.
[0003] In daily applications, video surveillance video has the characteristics of large amount of stored data and long storage time. Through the video, we can understand meaningful information such as pedestrians and vehicles that appear in a certain period of time, especially when the public security system is searching for clues and obtaining evidence. The traditional method will consume a lot of manpower, material resources and ti...
Examples
Embodiment Construction
[0018] A video summarization method based on moving object detection, the method includes: (1) obtaining an initial image sequence of an original video to be processed, performing moving object detection and analysis processing on the initial image sequence frame, and obtaining a moving object of each frame of the original video. Visual feature information, target tracking and merging processing is performed between frames, and the index information of each moving object in the original video is extracted; (2) According to the index information of each moving object in the original video, each moving object is reordered in time and space , and fuse the background images to generate a summary video, and record the index information of each moving object in the summary video; (3) According to the index information of each moving object in the original video and the index information of each moving object in the summary video, establish the moving object in each frame of the summar...