On-line learning tracking and engineering implementation method based on space planning

Abstract

The invention relates to an online learning tracking and engineering implementation method based on space planning, and belongs to the technical field of tracking algorithms. According to the method,an online learning thought is adopted, samples are updated in real time in the tracking process, a classifier is obtained through sample training, and candidate targets are detected; in order to counteract the boundary effect in the tracking process, a penalty factor is set for the algorithm to adjust the weight of a search area and is used for penalty of a correlation value of a background area so as to eliminate repetitive interference of the background area, and the real-time performance of the algorithm is improved by adopting a Gaussian iteration method. An SOT algorithm is constructed and implemented, the SOT algorithm is transplanted to a C66x series floating point type DSP processor, a DSP dual-core working mode is adopted, it is guaranteed that the algorithm has good tracking precision, and the tracking speed requirement of 50 frames per second is met.

Description

technical field [0001] The invention belongs to the technical field of tracking algorithms, and in particular relates to an online learning tracking and engineering realization method based on space planning. Background technique [0002] The purpose of tracking is to predict the trajectory of the target through the initial frame position information of the target in the image sequence. The tracker needs to obtain the apparent features of the target in a small number of training samples, and the extracted features are required to be robust to complex scenes such as target occlusion, fast motion and deformation. Therefore, the choice of the target appearance model is a key factor in determining the performance of the tracker. In a complex battlefield environment, there are a large number of tracking difficulties caused by human or non-human factors such as target occlusion, repetitive patterns, target rotation, and scale changes, and it is necessary to establish a robust tra...

AI Technical Summary

Problems solved by technology

However, there is a boundary problem in the previous algorithm, which produces repetitive interference in the background area, which affects the tracking performance of the algorithm.

Images