Visibility Detection Method of Traffic Condition Monitoring Video Based on Generating Confrontation Network
The traditional traffic condition monitoring video visibility detection method directly detects the most recognizable point of the human eye without lane line detection and interest region extraction,resulting in the poor detection effect of the traditional method.Therefore,the visibility detection method of traffic condition monitoring video based on generative confrontation network is proposed.On the basis of lane line detection and interest region extraction,the farthest point can be detected by human eyes based on the generate confrontation network,and further combined with computer vision and graphics,the visibility detection of traffic condition monitoring video is realized.The experimental results show that this method has better detection effect and can be applied widely.
NETL
NSTL
万方数据
