A Screening Method for On-orbit Aircraft Candidate Areas Based on Local Texture Density and Divergence

Abstract

The present invention provides a screening method for on-orbit aircraft candidate areas based on local texture density and divergence. A fixed-size sliding window is used to slide on the airport target candidate area SalMap. According to the difference between artificial ground objects and aircraft target textures, local texture Density removes objects with dense texture distribution, such as large areas of dense vegetation and building areas; considering that the aircraft is parked on the apron and runway with smooth texture, the texture of the aircraft area in the sliding frame is relatively concentrated and the divergence is small, and the divergence can be used Remove artificial objects with similar texture density and scattered texture distribution. Constructing dual-threshold logic judgment conditions through texture density and divergence, reducing the number of target candidate areas, improving the performance of false alarm rejection, and solving the poor screening effect of existing technology candidate areas. Too many artificial ground objects are divided into aircraft candidate areas. Thereby increasing the problem of subsequent data processing.

Description

technical field [0001] The invention belongs to the field of remote sensing image target detection, in particular to a method for screening candidate areas of on-orbit aircraft based on local texture density and divergence. Background technique [0002] It has always been an important research direction to develop space-based reconnaissance systems to use satellite remote sensing images to complete the detection of aircraft targets in the airport, so as to grasp the enemy's air force firepower equipment and military dynamics. Considering the limitation of satellite-to-ground data transmission bandwidth and the high real-time requirements of users to receive information, it is necessary to complete the rapid airport aircraft detection on the satellite, and only download the detection result data. As an early stage in the detection process, the degree of elimination of non-target areas determines the number of input objects for the subsequent complex false alarm elimination mo...

AI Technical Summary

Problems solved by technology

[0003] At present, in the airport aircraft candidate area extraction technology, part of the main focus is to obtain the airport runway information through the shape and texture feature detection of the airport runway detection, and then obtain the suspected aircraft candidate area. However, the shapes of different airport runways are not the same, and the aircraft may be parked on the apron. etc., such methods are not robust to

Another part of the method mainly uses the shape and gray information of the aircraft target to design the corresponding filter to filter the image, so as to obtain the position of the suspected target by using the corresponding threshold limit. However, it is only suitable for some models, and the effect is poor when the target contrast is low

Images