The invention discloses an intersection multi-path traffic flow artificial intelligence statistical method based on unmanned aerial vehicle aerial photography, which comprises the following steps: step 1, reading an aerial photography video shot by an unmanned aerial vehicle, taking a first frame image as a background, and manually marking four positions at each lane opening needing statistics to draw a marking frame R; establishing a statistical list Ln corresponding to each mark frame R; 2, processing each frame of image in the aerial photography video; step 2.1, detecting and tracking a vehicle in each frame of image; 2.2, judging the passing of each vehicle; step 2 and 3, traffic flow statistics; step 2, 3.1, counting the traffic flow of a single area; step 2, 3.2, carrying out path traffic flow statistics; and step 3, repeating the steps 2.1 to 2.3 until each frame of image of the aerial video is processed. According to the invention, classified statistics of multi-path traffic flow at a complex intersection is realized, the problem of singleness of traffic flow statistics paths is solved, and traffic flow statistics of complex paths is realized.
本发明公开了基于无人机航摄的路口多路径车流人工智能统计方法,包括如下步骤:步骤1、读取无人机拍摄的航摄视频,以并将第一帧图像为作为背景,人工在需要统计的每一车道口均标记四个位置绘制标记框R;并建对应每一标记框R均立统计列表Ln;步骤2、对航摄视频中每一帧图像进行处理;步骤2.1、对每一帧图像中的车辆进行检测与跟踪;步骤2.2、每一车辆的经过判断;步骤2,3、车流量统计;步骤2,3.1、单个区域车流量统计;步骤2,3.2、路径车流量统计;步骤3、重复步骤2.1至2.3直至完成航摄视频每一帧图像的处理。本发明实现复杂路口中多路径车流的分类统计,解决车流统计路径的单一性问题,实现复杂路径的车流统计。
Intersection multi-path traffic flow artificial intelligence statistical method based on unmanned aerial vehicle aerial photography
基于无人机航摄的路口多路径车流人工智能统计方法
2024-08-16
Patent
Electronic Resource
Chinese
Unmanned aerial vehicle aerial video traffic flow statistical system
European Patent Office | 2024
|