Multi-camera-based multi-objective positioning tracking method and system

Abstract

The invention discloses a multi-camera-based multi-objective positioning tracking method. The method is characterized by comprising the following steps: installing a plurality of cameras at a plurality of visual angles firstly, planning a public surveillance area for the cameras, and calibrating a plurality of height levels; sequentially implementing the steps of foreground extraction, homography matrix calculation, foreground likelihood fusion and multi-level fusion; extracting positioning information which is based on selected a plurality of height levels and obtained in the step of foreground likelihood fusion; processing the positioning information of each level by using the shortest path algorithm so as to obtain the tracking paths of the levels; and after combining with the processing results of foreground extraction, completing the multi-objective three-dimensional tracking. By using the method disclosed by the invention, in the process of tracking, the vanishing points of the plurality of cameras are not required to be calculated, and a codebook model is introduced for the first time for solving the multi-objective tracking problem, thereby improving the accuracy of tracking; and the method has the characteristics of good stability, good instantaneity and high precision.

Description

technical field [0001] The invention relates to a multi-camera-based multi-target positioning and tracking method and system. Background technique [0002] Multi-target tracking is a hot research issue in the field of computer vision. It is the premise and basis for understanding the behavior of moving targets. It has a wide range of applications in robot navigation, video understanding, intelligent monitoring and other fields. In the specific implementation process of multi-target tracking, the most difficult thing to deal with is that multiple targets frequently occlude each other. In this case, one foreground region may belong to multiple objects. According to the traditional color distribution, shape and other information, it is almost impossible for us to achieve accurate monitoring and tracking of moving targets. In recent years, many scholars have done research in this area, and the proposed methods can be roughly divided into two categories: single-camera tracking ...

AI Technical Summary

Problems solved by technology

[0006] However, the above method uses a mixed Gaussian model to model the background, which often produces holes and shadows, and the failure of target monitoring will have a great impact on subsequent tracking.

This method needs to calculate the hidden point of each camera to obtain the homography matrix between multiple views. The calculation of the hidden point is a relatively complicated process, and the calculation error of the hidden point will lead to the calculation of the homography matrix. Error, which leads to the failure of multi-target tracking

Another problem is that this method uses the graph cut theory to process the positioning information of the 10th floor of the space to achieve tracking, which makes this method unable to meet the real-time requirements at all.

These three significant shortcomings greatly limit its application in engineering

Images