Double-load remote sensing image fusion method

Abstract

The invention discloses a double-load remote sensing image fusion method. The method includes: 1, collecting an infrared image and a visible light image of a same scene at the same time; 2, acquiringbackground mean value parameter value of the visible light image; 3, performing target detection on the visible light image to acquire a target slice; 4, acquiring physical position of a target according to image coordinates of the target slice; 5, acquiring image coordinates of the target on the infrared image according to the physical position of the target, and cutting out an image block F containing the target by taking the coordinates as a center; 6, setting a threshold value P according to the parameter value, and counting the number N of pixel points with pixel point grayscale value greater than P in the image block F; when the number N is within a set numeric value interval and the pixel points greater than P are in centralized distribution, judging the target slice as a real target, and when the number N is greater than a right-end value of the set numerical value interval, judging the target slice as a real target. By the method, false alarm rate in optical image target detection can be lowered, and target detection rate can be increased.

Description

Technical field [0001] The invention belongs to the technical field of remote sensing image detection, and specifically relates to a dual-load remote sensing image fusion method. Background technique [0002] Due to the complex carrier environment, the target information has obvious diversity and uncertainty. However, the information provided by a single remote sensing information source has limitations and it is difficult to obtain an accurate description of the scene. The comprehensive use of high-resolution visible light, infrared images and other information to achieve accurate detection and recognition of time-sensitive targets (aircraft, ships, vehicles, etc.) is an important development direction for high-resolution image target detection and recognition technology. In the current syn-optical integrated satellite, the visible light and infrared imaging equipment are on the same optical axis, which provides the basis for the fusion of the two types of images. [0003] In inf...

AI Technical Summary

Problems solved by technology

[0003] In infrared images, targets such as ships, airplanes, vehicles, etc. have obvious contrast with the background area, and the outline information of the target is more prominent, while in visible light, the target is easily disturbed by painting and background, which is not conducive to the extraction of target candidate areas; , the time-sensitive target has strong saliency in the local background, that is, the target has a large difference from the background in the local range

Visible light images have rich target information, precise contours, and obvious texture features, but they cannot work at night and have no camouflage recognition capabilities; infrared data has the advantages of irrelevant lighting conditions, uninterrupted observation, high measurement accuracy, and strong target recognition capabilities. Vague

Images