A local data association method for moving vehicle tracking in aerial video

Abstract

The invention discloses a local data association method for tracking a moving vehicle in an aerial video. According to the vehicle motion information in the video frame, the area of ​​interest of the vehicle to be tracked is determined, and the range of data association in subsequent video frames is narrowed; The motion features, appearance features, and shape features of the vehicle and the detection vehicle in the region of interest; respectively establish the motion, appearance, and shape feature cost functions, and fuse the three to form a comprehensive cost function between the vehicle to be tracked and the detection vehicle; according to Four kinds of matching relationships between the vehicle to be tracked and the vehicle detected in the region of interest in the subsequent frame, and four different data association strategies are formulated to flexibly and effectively solve the problem that the vehicle enters and exits the image. When the present invention performs data association, Only considering the vehicles detected in the ROI can make up for the defect that the number of vehicles in the front and rear frames of the global data association algorithm needs to be fixed, and can significantly improve the accuracy and efficiency of moving vehicle data association.

Description

technical field [0001] The invention relates to the technical field of vehicle tracking, and proposes a local data association method for moving vehicle tracking in aerial video. Background technique [0002] Moving vehicle tracking in UAV aerial video is an important part of intelligent transportation system construction, and it is also an important content of target motion analysis research. The purpose of vehicle tracking is to find and extract moving vehicles in real time in a continuous image sequence, and continuously track them according to the vehicle's edge, local motion, and grayscale change information to obtain the movement trajectories of these moving vehicles, which can be used for the next step of vehicle tracking. Identification provides data, and can also provide important references for traffic flow analysis and intelligent transportation system construction. [0003] For moving vehicle tracking, the core problem is data association, which is to associate ...

AI Technical Summary

Problems solved by technology

However, the disadvantage of the joint probability data association algorithm is that the calculation of the association probability is complex and time-consuming, and it is difficult to meet the real-time requirements of practical problems.

Moreover, it only considers position or motion information during association, and does not consider other possible information, such as appearance, shape information, etc.

[0004] The currently commonly used global data association algorithm mainly has the following weaknesses: first, it has a large amount of calculation and low efficiency; second, it is only applicable to the situation where the number of vehicles in the front and rear frames is fixed; third, it does not make full use of the characteristics of vehicles.

The specific performance is that when performing data association, the vehicle to be tracked is often associated with all detected vehicles in subsequent frames, resulting in a large amount of calculation; and the number of vehicles in the front and rear frames must be fixed, which is poor in flexibility; The impact of the vehicle's appearance features and shape features on the tracking effect is ignored during tracking, and the robustness is poor

Therefore, the tracking effect is poor during vehicle tracking, and even tracking errors occur

Images