Concentrated video constructing method based on irregular target boundary object

Abstract

The invention relates to a concentrated video constructing method based on an irregular target boundary object. The concentrated video constructing method based on an irregular target boundary objectextracts the target in a monitoring video, rearranges the target and displays the target on a fixed background picture in a convenient, quick and efficient mode, uses the irregular boundary expressionmode for the detected target in the original video, instead of using a traditional rectangle frame expression mode. Therefore, the concentrated video constructing method based on an irregular targetboundary object greatly improves the target tracking accuracy of the concentration system and the overall fusion effect of the concentrated video, and can obtain the benefits without increasing systemburden obviously.

Description

technical field [0001] The invention relates to a condensed video construction method based on irregular target boundary objects, belonging to the technical field of digital video. Background technique [0002] With the development of full-scale digital and networked video surveillance systems, the role of video surveillance has become more and more obvious. Its high degree of development, integration and flexibility provides a broader development space for the development of the entire security industry. Intelligent video surveillance has become a new generation of video surveillance trend due to its more intelligent, active, and effective features. The demand for intelligent video surveillance systems mainly comes from occasions that are sensitive to security requirements, such as the military, public security, banks, roads, parking lots, etc. When theft or abnormality occurs, this type of system can proactively issue an alarm to the security personnel in a timely and acc...

AI Technical Summary

Problems solved by technology

For surveillance videos in relatively lively areas, there are moving objects in almost every frame, and the summary video cannot change the time sequence between objects. Therefore, it is difficult to achieve the effect of speeding up browsing and reducing storage by using summary videos. Therefore, we focus on Placed on condensed video research

[0007] Video condensing also has many deficiencies at this stage: (1) The detection rate and correct rate of target detection are still difficult to achieve perfection; (2) Sometimes there are problems of missing and wrong tracking in the process of target tracking; (3) The condensed video Results may show localized tonal differences

Inaccurate target tracking and local tone differences in condensed video results are directly caused by the use of rectangular boxes to represent targets. To solve these two problems, we must start from the root

Images