Aircraft target detection method based on PolSAR image multi-feature fusion

Abstract

The invention discloses an aircraft target detection method based on PolSAR image multi-feature fusion. The aircraft target detection method comprises the steps of: preprocessing a PolSAR image to obtain a filtered and deoriented coherence matrix; calculating a scattered power value of each pixel point, performing region segmentation on the PolSAR image by using the scattered power values, and segmenting a region of interest with the result being a binary image; performing regional screening on the region of interest to obtain a binary image of a suspected aircraft target; performing a morphological expansion processing on the binary image of the suspected aircraft target to obtain a binary image of a background region around the suspected aircraft target; acquiring deoriented coherence matrixes corresponding to the suspected aircraft target and the background region, and calculating features to be detected of the suspected aircraft target; and further screening the features to be detected to obtain a final detection result. The aircraft target detection method can reduce the computational efficiency through performing coarse extraction on the PolSAR image, takes the aircraft target as a region for processing, and can fully utilize the polarization property of the overall aircraft target.

Description

technical field [0001] The invention belongs to the technical field of PolSAR image interpretation, in particular to an aircraft target detection method for PolSAR image multi-feature fusion. technical background [0002] As an important remote sensing device, Polarimetric Synthetic Aperture Radar (PolSAR) has been widely used in military and civilian fields for its all-time, all-weather, high-resolution and strong penetration characteristics. Compared with the rapid development of PolSAR system research and development, the research on PolSAR image interpretation is still relatively lagging behind. Under the trend of massive growth of various remote sensing data, remote sensing image interpretation is particularly important. As an important part of image interpretation, aircraft target detection has important military and national defense significance. For aircraft target detection, some existing methods need to traverse all the pixels in the image; some require certain tr...

AI Technical Summary

Problems solved by technology

For aircraft target detection, some existing methods need to traverse all the pixels in the image; some require certain training samples for subdivision; others need to extract the airport runway area in advance, and then traverse all the airport areas In addition, the main problem of aircraft target detection in low-to-medium resolution and complex large-scene SAR images is that the shape characteristics of the aircraft cannot be reflected, and can only be expressed as some pixel features. Common statistical signatures are no longer applicable, making it more difficult to detect aircraft targets

Images