Adaptive segmentation method of SAR (Stop and Reveres) image water area

Abstract

The invention discloses an adaptive segmentation method of an SAR (Stop and Reveres) image water area, which mainly solves the problems of fixed parameters and inaccurate division boundaries existing in the prior art. The method comprises the following steps of (1) manually obtaining a label set and establishing a corresponding mathematical model; (2) obtaining a first segmentation result by utilizing the mathematical model; (3) adaptively obtaining a target area of secondary segmentation according to the first segmentation result; and (4) automatically generating labels in the target area, and computing a geodesic distance to obtain a final segmentation result. The invention can adaptively compute the target area of the secondary segmentation according to the boundary of the first segmentation, reduce the computing complexity resulted by using fixed bandwidth parameters, simultaneously improve the accuracy of the secondary segmentation, and can be used for SAR image water area segmentation, water area detection and bridge detection.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a method for segmenting SAR image water areas, which can effectively and adaptively select target areas for secondary segmentation. Background technique [0002] SAR image water segmentation has become an increasingly important tool in environmental monitoring, such as flood monitoring and coastline extraction. [0003] Compared with other types of optical satellite images, SAR images have obvious advantages due to their all-weather and all-time characteristics. For example, due to the outstanding feature that SAR images can be shot through clouds, this provides an important basis for flood monitoring, because areas where floods occur usually have thicker clouds above them. SAR image water segmentation has its own problem characteristics. The smooth water surface is equivalent to a specular reflector for the incident wave in the radar band. It absorbs almost most of the ...

AI Technical Summary

Problems solved by technology

Because the SAR image itself has coherent speckle noise interference, and the background around the water area is very complex, this makes SAR image water segmentation difficult.

Many existing SAR image water segmentation methods rely on manual selection of key parameters, and the segmentation results are unstable due to parameter selection. For example, the water extraction method based on the degree of evacuation proposed by Shan Huang et al. (APSAR2009, Xian, China. 2009: 498- 501) and the SAR image water area recognition method based on grayscale statistics and area coding proposed by Yang Lei et al. (patent application number: 200810236455)

When these methods are applied to images with large size and complex background, the results are often not ideal and the processing time is longer

Images