Target tracking algorithm based on object space relationship

Abstract

The invention relates to a target tracking algorithm based on an object space relation, and the algorithm specifically comprises the following steps: G1, taking the features of an image and the position of a target in the image as the input, capturing the correlation of the targets in two frames of images through employing a five-layer convolution small network, and finally obtaining a feature similarity matrix between the targets; and G2, obtaining the distance between the center points of the targets to serve as a spatial similarity matrix between the targets, performing Hadamard product onthe spatial similarity matrix and the feature similarity matrix, calculating the relevance of the targets, preventing the targets with high apparent feature similarity and relatively far spatial position distance from being wrongly associated, and finally obtaining a target tracking result. The problem that the tracking effect of a target tracking algorithm only utilizing the target apparent characteristics is reduced in a scene with high target apparent characteristic similarity and low spatial characteristic similarity can be solved.

Description

technical field [0001] The invention relates to the technical field of multi-target tracking, in particular to a target tracking algorithm based on object space relationship. Background technique [0002] In essence, artificial intelligence is simulating the process of human processing problems. In the human perception system, the perception of the target includes not only the apparent features of the target (such as color, texture, shape, etc.), but also the spatial position of the target and the relative relationship between the targets. Based on these apparent and spatial features, humans can easily tell whether two targets are the same object. Although the current multi-target tracking algorithms use the powerful information capture ability of convolutional neural networks to achieve good tracking results, from the perspective of the human perception system, most multi-target tracking algorithms only simulate the human perception of the apparent features of the target. ...

AI Technical Summary

Problems solved by technology

[0006] However, the spatial relationship between targets cannot be ignored for target tracking. It is relatively one-sided to determine the correlation between targets only based on the similarity of target apparent features. When the apparent feature vectors of two targets in the image are similar, A tracking algorithm that only uses the apparent characteristics of the target will cause confusion and easily lead to tracking errors. For example, when a certain target has been stationary at the lower right of the image, but the algorithm mistakenly believes that it is related to a target at the upper left of the image, so how to consider The role of target appearance features and spatial relationships in target tracking is particularly important to further improve the effect of target tracking

Images