Everyone a great god, excuse me for a number of different Angle of view camera approximate the video from the same place, how to effectively extract the future goals? Specific how to program? First to background modeling background in video sequences, and then through the background deduction to extract the target foreground image, and prospect target through morphology processing, connectivity analysis, image binarization processing eventually moving targets, the final matching to extract the features for the next goal, specific how to code? O share code

CodePudding user response:

Baidu search relevant keywords,

CodePudding user response:

I want to know a specific implementation process, rather than the sort of pure theoretical papers, want to know a general idea of solving the problem

CodePudding user response:

https://www.baidu.com/s? Ie=utf-8 & amp; F=8 & amp; Rsv_bp=0 & amp; Rsv_idx=1 & amp; Tn=baidu& Wd=20 matlab source code % % 20 opencv % 20 background deduction to extract the target foreground image