A method and device for automatic water segmentation of SAR images with complex terrain

Abstract

The invention discloses a water area automatic segmentation method for a SAR image with complex terrain, comprising: preprocessing the original SAR image to obtain a preprocessed SAR image; using a threshold segmentation method to perform the first water area segmentation on the preprocessed SAR image , to obtain the first water area; extract the ROI in the first water area; use the LGDF ACM algorithm to perform a second water segmentation on the ROI to obtain the second water area. The invention also simultaneously discloses a water area automatic segmentation device for SAR images of complex terrain. By adopting the technical scheme of the invention, the automatic segmentation of the SAR image water area can be realized under complex terrain.

Description

technical field [0001] The present invention relates to image processing technology in Synthetic Aperture Radar (SAR, Synthetic Aperture Radar), in particular to a water area automatic segmentation method and device for SAR images with complex terrain. Background technique [0002] Continuous quantitative fixed-point monitoring of large areas of water is related to the balance of water and soil and the protection of the economic environment. Especially in the face of natural disasters, such as earthquakes and floods, water monitoring is of great significance for disaster prevention and prediction. Because SAR has the advantages of imaging the observation area all day, all weather, and at a long distance, SAR is often used to observe waters and calculate the area of ​​waters. Through multiple observations of the same waters at different times, the area of ​​waters can also be obtained. The difficulty and core of the change information is to extract the water area accurately a...

AI Technical Summary

Problems solved by technology

However, most of the methods are based on the assumption of a flat surface. Under complex terrain conditions, SAR images will be affected by mountains and other ground objects and produce shadows. The echo amplitude is equivalent to that of water, which directly leads to the current commonly used algorithms, such as The segmentation effect of algorithms such as segmentation and level set segmentation is greatly affected, or even invalid

Images