High-resolution single-polarization SAR image building target detection method based on structural characteristics

Abstract

The invention provides a high-resolution single-polarization SAR image building target detection method based on structural characteristics, and the high-resolution single-polarization SAR image building target detection method comprises the steps: inputting a high-resolution single-polarization SAR image, carrying out image segmentation, and obtaining a probability value prob that SAR image pixels belong to a building target and a background through a deep semantic segmentation network; performing intersection operation by using the region number and the connected domain number to obtain an over-segmentation result of the image; calculating a region adjacency map, performing similarity analysis on adjacent regions according to the region adjacency relationship, the region geometrical characteristics, the statistical characteristics, the texture characteristics and the obtained probability value prob, combining the regions meeting a similarity threshold value, sequentially expanding and combining outwards, and calculating the adjacency map of each region of the image after the regions are combined; and based on different image features represented by shadows of the roof, the wall body and the ground of the building target in the SAR image, judging whether the preliminarily extracted area and the adjacent area can form a corresponding structure of the building target or not by utilizing the area adjacent graph, and obtaining a building target detection result.

Description

technical field [0001] The invention belongs to the technical field of synthetic aperture radar (Synthetic Aperture Radar, SAR) image target recognition, and further relates to a high-resolution single-polarization SAR image building target detection method based on structural characteristics in the field of SAR image target recognition. Background technique [0002] Microwave remote sensing is an important branch of remote sensing technology. Among them, SAR, as a microwave imaging radar, can monitor ground activities in real time without being affected by clouds and weather, and has attracted widespread attention. SAR images play an important role in tasks such as urban monitoring, urban planning, and urban disaster analysis. Therefore, the extraction of architectural objects in SAR images has attracted people's attention. [0003] In high-resolution SAR images, due to the special imaging mechanism, there will be effects such as overlapping, dihedral reflection, and shadow...

AI Technical Summary

Problems solved by technology

However, the imaging mechanism of SAR is different from that of optics. Interference such as coherent speckle and complex background environment, and the target characteristics under different imaging conditions and parameters are quite different. These problems bring huge challenges to the detection method of building targets based on deep learning. Learning object detection methods or semantic segmentation algorithms cannot achieve optimal performance in high-resolution SAR images

[0005] In summary, with the development of SAR imaging technology, its resolution has been continuously improved, and the details of architectural objects can be reflected, but at the same time it also presents complex and diverse structural characteristics

Relying on the statistical features, texture features and specific features of building objects is difficult to accurately model single buildings in high-resolution SAR images

Images