ROI (Region of Interest) extraction method of ground target of unmanned aerial vehicle

Abstract

The invention discloses an ROI extraction method of a ground target of an unmanned aerial vehicle. The ROI extraction method of the ground target of the unmanned aerial vehicle comprises preprocessing a reference satellite image and extracting a mode which can reflect local features of a target; performing statistical analysis on mode distribution condition of a target region and an associated environment region, calculating a saliency value of every mode, and establishing a target visual saliency model; for a real-time image, obtaining the mode of every pixel point and a corresponding saliency value of the mode of every pixel point, wherein the saliency value corresponding to the mode of every pixel point corresponds to the visual saliency image of the real-time image; estimating a target rough position of the visual saliency image; estimating rough resolution of the real-time image with the target rough position as the center and by utilizing size distribution information of modes in a region with a size identical with that of the target region; obtaining an ROI basically identical with the resolution of the reference satellite image according to the target rough position and an estimated resolution value.

Description

technical field [0001] The invention belongs to the technical field of image target detection and recognition, and in particular relates to a method for extracting an ROI of an unmanned aerial vehicle to a ground target. Background technique [0002] Specific target detection technology in aerial images is of great significance in the monitoring of specific areas in the battlefield and precise strikes on targets. Due to the limitation of objective conditions, it is difficult to obtain reference images of specific targets in many cases. At present, the commonly used method is to use satellite optical images or aerial images captured by high-altitude reconnaissance aircraft as the reference image of a specific target, and use the image captured by the airborne camera during the flight as the real-time image, and complete it in the real-time image through the target detection algorithm. Detection and localization of specific objects, such as figure 1 shown. The main difficul...

AI Technical Summary

Problems solved by technology

(2) The load of small UAVs is limited, and there is carrier jitter. From the perspective of algorithm stability, it is generally required to run on DSP

The main problems of this type of method are: the large number of real-time image features makes the search range of feature matching very large, and the extraction of features with scale or affine invariance also requires high computational complexity, which is difficult to meet the requirements of onboard processing. needs

However, the current ROI positioning methods are all based on bottom-up thinking, without using the information of the target itself, the extracted ROI may miss the ta

Images