Video-based multi-target pedestrian detection and tracking method

Abstract

The invention discloses a video-based multi-target pedestrian detection and tracking method. A YOLO3 target detection algorithm with good speed and accuracy is utilized, video images under different scenes are constructed, and a detection model is trained, so that the influence of illumination change and visual angle change is overcome, and efficient detection of multi-target pedestrians is ensured; the multi-target pedestrian is effectively tracked based on the Kalman filtering algorithm and the Hungary algorithm effectively tracks, the problem that repeated target detection often exists in multi-target detection is avoided, thereby realizing the multi-target pedestrian tracking method taking the Deep-SORT algorithm as the core. The method has the characteristics of efficient multi-targetpedestrian detection and efficient multi-target pedestrian tracking.

Description

technical field [0001] The present invention relates to a computer vision system, more specifically it relates to a computer vision-based detection and tracking method. Background technique [0002] Pedestrian detection and tracking refers to the process of detecting the position information of pedestrians from video sequences, and continuously tracking the moving pedestrian targets to determine their trajectory. Pedestrian detection is the basis and premise of pedestrian tracking, so a good pedestrian detection algorithm can provide strong support and guarantee for the latter. Pedestrian detection belongs to the category of target detection. In recent years, thanks to the development of target detection technology, the multi-target tracking method based on detection has become the main method of multi-target tracking. The tracking process of the detection-based multi-target tracking method can be transformed into a data association problem. This method is highly dependent ...

AI Technical Summary

Problems solved by technology

The tracking process of the detection-based multi-target tracking method can be transformed into a data association problem. This method is highly dependent on the detection results, and complex backgrounds and occlusions between targets will have a greater impact on target detection, which in turn will cause data association problems. At the same time, the target model relied on to establish an accurate correspondence between multiple detection values ​​and multiple tracking values ​​will also greatly affect the effect of multi-target tracking

Images