A target positioning and tracking method and system based on multi-band information fusion

Abstract

The invention relates to a target positioning and tracking method and system based on multi-band information fusion, comprising: collecting an infrared light image sequence and a visible light image sequence of a target; extracting image features corresponding to the infrared light image sequence and the visible light image sequence respectively and the target suspicious area; fuse all the extracted image features with the target suspicious area and determine the target area; select the target to be tracked in the target area; obtain the target according to the position information of the target to be tracked and the servo angle of the tracking frame absolute position; performing track fitting based on the target absolute position to obtain a tracking frame servo angle update value; updating the angle of the servo tracking frame based on the tracking frame servo angle update value. The invention improves the speed and accuracy of target detection and recognition, as well as the accuracy and response speed of target tracking, and can also improve the robustness and anti-interference ability of the system.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a target positioning and tracking method and system based on multi-band information fusion. Background technique [0002] At present, the main purpose of computer vision is to replace human eyes with cameras and central processing units to identify, track and measure objects, so as to build artificial intelligence systems that can obtain "information" from images or multi-dimensional data. The target detection and tracking algorithm model represented by correlation filtering and convolutional neural network and the reinforcement learning represented by deep neural network continue to develop, and are gradually applied to face recognition, unmanned driving, intelligent robot, unmanned hotel, unmanned Airports, unmanned supermarkets, etc. [0003] In the related art, on the one hand, the target detection method cannot achieve both speed and accuracy. To achieve higher target detecti...

AI Technical Summary

Problems solved by technology

[0003] In related technologies, on the one hand, the target detection method cannot achieve both speed and accuracy. To achieve higher target detection and recognition accuracy, a deeper and more complex convolutional network is required, which brings a greater amount of calculation.

On the other hand, there are more or less problems in the target tracking method that the target is lost during long-term tracking and the target gate cannot be automatically increased or decreased according to the actual scene. These problems have restricted the application and development of computer vision to a certain extent.

Images