Target recognition method and device based on visible light and infrared multispectral image sequence

Abstract

The invention provides a target recognition method and device based on visible light and infrared multi-spectral image sequences, wherein the method includes: using a multi-spectral camera to collect multi-spectral image sequences of multiple bands at uniform intervals for a target area; using a near-infrared camera at the same time , mid-infrared cameras and far-infrared cameras respectively collect infrared image sequences corresponding to each infrared camera in the corresponding spectral range; at the same time, visible light cameras are used to capture color image sequences; motion-based foreground extraction algorithms and appearance-based foreground extraction algorithms are used respectively. The foreground extraction algorithm obtains the target synchronous image containing the candidate rectangular window; recognizes the target category recognition probability contained in the candidate rectangular window according to the pre-trained classifier corresponding to each target synchronous image; displays the weighted fusion result of the target category recognition probability. Therefore, the characteristics of the target in various dimensions are fully utilized, and the accuracy of target recognition can be greatly improved.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a target recognition method and device based on visible light and infrared multispectral image sequences. Background technique [0002] In recent years, the development of artificial intelligence and computer vision has greatly improved the performance of image / video-based target detection and recognition methods, making it a new method to study such problems. Compared with methods based on radar and radio frequency, the methods based on machine vision have the advantages of simple system, small hardware size, strong scene universality, long detection distance, and fine-grained recognition. These characteristics also make the image / video-based recognition method applied in a variety of real-world scenarios, such as subway security detection temperature and people flow statistics, vehicle automatic driving, drone inspection, commodity shelf sorting, etc. These practical...

AI Technical Summary

Problems solved by technology

However, these works only meet one or a few types of recognition scenarios, and the recognition accuracy is low, it is difficult to deal with environmental changes, complex backgrounds, and multiple instances, and cannot provide high-precision recognition results for most common recognition scenarios.

Images