Correlation Filtering Moving Target Tracking Method Based on Memory Mechanism and Convolution Feature

Abstract

The invention proposes a correlation filter moving target tracking method based on memory mechanism and convolution feature, which belongs to the technical field of computer vision. The method of the present invention utilizes the pre-trained deep convolutional neural network to extract the convolution features of the target, inspired by the memory mechanism of the human brain in the cognitive behavior of human visual information processing, and integrates the memory mechanism into the detection and detection of the classifier of the correlation filtering method. During training and updating. Among them, the memory mechanism consists of three parts: response map decision, adaptive peak detection and adaptive fusion coefficient. The method of the present invention has strong robustness, and can still continuously and stably realize target tracking under conditions such as severe deformation of the target, reappearance after disappearing for a short time, or occlusion. At the same time, it has a high target tracking speed, reduces the complexity, and reduces the amount of calculation.

Description

technical field [0001] The invention relates to a method for tracking a moving target in an image sequence, in particular to a method for tracking a moving target with correlation filtering based on a memory mechanism and convolution features, and belongs to the technical field of computer vision. Background technique [0002] Moving object tracking technology is an important research direction of computer vision science, and it is widely used in security monitoring, human-machine interface, medical diagnosis and other fields. At present, the main problem of moving target tracking technology is that it is difficult to overcome the influence of complex interference factors such as changes in background lighting conditions, occlusion of targets, shape changes, size changes, and fast movements, resulting in a decrease in tracking accuracy. [0003] The discriminative tracking method is an important moving target tracking method, including: multiple sample learning (Multiple Ins...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to solve the problem of accurate and high-speed tracking of the target under interference conditions such as sudden change of attitude and shape of the target, reappearance after short-term disappearance, and occlusion, and propose a correlation filtering moving target based on memory mechanism and convolution features The tracking method integrates the memory mechanism of the human brain into the classifier detection, training and updating process of the correlation filtering algorithm, which can realize moving target tracking with high precision, strong robustness and fast operation speed in complex application scenarios

Images