The scene of surveillance video is usually wide and variable, and the target vehicles inside it are often small. Surveillance videos usually are not suitable for intelligent video applications, such as video vehicle detection. To solve this problem, an extraction method of road structure for surveillance video is proposed for intelligent applications. The principle of the method is developed to first extract road regions of surveillance videos by accumulating movement regions in video image sequences. Based on it, the left and right borderlines of road regions are pulled out using the Hough transform according to the statistical characteristics of road structure in surveillance video. Experiment results have shown that the road structure can be extracted precisely and effectively from the surveillance video of freeway and city expressway.
An Extraction Method of Road Structure for Surveillance Video
11th International Conference of Chinese Transportation Professionals (ICCTP) ; 2011 ; Nanjing, China
ICCTP 2011 ; 1533-1542
2011-07-26
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
VIDEO SURVEILLANCE EQUIPMENT, VIDEO SURVEILLANCE SYSTEM, AND VIDEO SURVEILLANCE METHOD
Europäisches Patentamt | 2017
|ROAD SURVEILLANCE SYSTEM, ROAD SURVEILLANCE METHOD, AND NON-TRANSITORY STORAGE MEDIUM
Europäisches Patentamt | 2024
|