Intelligent target tracking and loss re-detection method based on hierarchical perceptron

Abstract

The invention relates to an intelligent target tracking and loss re-detection method based on hierarchical perceptron, which comprises the following steps: firstly, calculating feature maps of images of two different search areas according to the position and the size of a selected tracking target, and collecting samples as training samples of two different perceptron; when a tracked first frame image is input, enabling the perceptron to complete initialization training and initialize a target state, a position and a reference response value; when images of a second frame and a subsequent frame are input, judging whether image data of a current frame enter a tracking branch or a re-detection branch; when entering a tracking branch, calculating the target position of the current frame by using two tracking branches with hierarchical structures established by a multi-stage sensor, outputting the optimal target position after comparison, and judging the target state; when entering a re-detection branch, enabling a re-detector composed of a plurality of primary perceptron to re-detect a target in a large range, and using the maximum response value area as a re-detection target and judging the state of the target at the same time.

Description

technical field [0001] The invention relates to the field of intelligent video processing, in particular to an intelligent target tracking and loss re-examination method based on a hierarchical perceptron. Background technique [0002] The task of object tracking is to continuously locate the location of the object in subsequent video frames by using the location information of the object selected in the initial frame of the video. Object tracking technology has a wide range of applications in human-computer interaction, automatic driving and security monitoring. In traditional optoelectronic platforms, targets mainly rely on manual control of the platform for tracking and continuous observation. In the modern optoelectronic platform, the intelligent target tracking method is responsible for continuously outputting the position information of the locked target to control the direction and pitch angle of the optoelectronic platform, so that the target is always located in th...

AI Technical Summary

Problems solved by technology

[0003] Problems solved by the present invention: Aiming at the problems of low tracking accuracy, slow calculation speed and inability to re-detect lost targets when target tracking algorithms are applied to modern optoelectronic platforms, the present invention provides an intelligent target tracking and tracking system based on hierarchical perceptrons. Lost re-examination method to achieve a large range of targets (padding=6, padding=6 refers to the area six times the size of the target

Images