Infrared small target detection method based on multi-direction morphological filtering complex cloud background

Abstract

The invention discloses an infrared small target detection method based on a multi-direction morphological filtering complex cloud background. A multi-direction Tophat method is provided for filtering according to the difference of the geometrical morphology characteristics after cloud layer edge and target imaging, and a target segmentation threshold is determined by the gray difference between an original image and a filtering image so as to extract an infrared target. According to the infrared small target detection technology, a multi-direction filtering module based on the morphology is designed according to the geometrical morphology after cloud layer edge imaging, the Tophat filtering method, mainly including median filtering, corrosion and expansion filtering methods, is applied to extract background information covering the cloud layer edge apart from the infrared small target, and the target is extracted through the segmentation threshold generated by the difference value image.

Description

Technical field [0001] The present invention relates to background preprocessing technology and target extraction technology, in particular to a small target detection method applied to an infrared imaging system. Background technique [0002] In modern high-tech warfare, in order to be able to detect enemy satellites, incoming missiles, aircraft and other military targets as soon as possible, so that the guidance system has sufficient reaction time, infrared reconnaissance systems are required to be able to detect targets at a long distance. Only by discovering, tracking, capturing and locking the target in time, can it be effectively intercepted or attacked. Compared with the large-area sky background, the target only occupies one or a few pixels in the infrared image, and because the target image is small, it is easily blocked by clouds during the movement, which causes great interference to the detection of infrared targets. [0003] At this stage, the detection algorithms for...

AI Technical Summary

Problems solved by technology

Compared with the large-area sky background, the target only occupies one or a few pixels in the infrared image, and because the target image is small, it is easily blocked by clouds during the movement process, causing great interference to the detection of infrared targets

[0003] At this stage, the detection algorithms based on complex cloud backgrounds for infrared imaging systems mainly include time-domain high-pass filtering methods, genetic algorithm theory, artificial neural network methods, fractal theory and other target extraction methods. Such methods have high algorithm complexity and poor real-time performance. Satisfy the rapid response process of the infrared imaging system, and it is difficult to realize in engineering

Images