Target tracking method and system combining channel selection and position optimization

Abstract

The invention discloses a target tracking method and system combining channel selection and position optimization, and relates to the field of target tracking, and the method comprises the steps: obtaining a to-be-tracked video sequence; performing feature extraction by using a CNN network model according to the video sequence to be tracked to obtain convolutional features; the convolution features comprise a third convolution feature, a fourth convolution feature and a fifth convolution feature; performing gradient calculation, channel selection and cross-correlation operation in sequence according to the convolution features to obtain a first target response diagram; determining a target initial positioning position and a peak sidelobe ratio according to the first target response diagram; and performing position optimization according to the initial positioning position of the target and the peak sidelobe ratio to obtain a final positioning position of the target. According to the method, the tracking precision and speed are considered, and the problem of tracking drift in scenes such as large deformation or low resolution of the target is solved.

Description

technical field [0001] The invention relates to the field of target tracking, in particular to a target tracking method and system combining channel selection and position optimization. Background technique [0002] Object tracking is one of the research hotspots in the field of computer vision, and it has a wide range of applications in human-computer interaction, medical imaging, traffic monitoring and other fields. The basic process of object tracking is to mark the object bounding box in the first frame of the video sequence to be tracked and accurately locate the object in the subsequent frames. Due to the fact that the target object may be deformed, rotated, and affected by many factors such as illumination changes in the natural environment, there are still many problems to be solved in target tracking. [0003] In recent years, deep learning methods have emerged and have been widely used in the field of target tracking. Trackers such as DeepSRDCF, CF2, ECO, STRCF, a...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a target tracking method and system combining channel selection and position optimization, taking into account the tracking accuracy and speed, and solving the problem of tracking drift in the scene of large deformation or low resolution of the target

Images