SAR image denoising method based on multi-scale cavity residual attention network

Abstract

The invention relates to an SAR image denoising method based on a multi-scale hole residual attention network. The method comprises the following steps of by extracting features of different scales ofthe image through a multi-scale convolution group, broadening a convolution kernel receptive field by utilizing cavity convolution; exuecting more context information of the image; and transmitting the feature information of the shallow layer to a deep convolutional layer by using jump connection to keep image details, adding an attention mechanism to intensively extract noise-related features, and automatically learning the SAR image speckle noise distribution form in combination with a residual error learning strategy to achieve the purpose of removing speckle noise. Experimental results show that compared with a traditional SAR image noise removal method, the method has the advantages that the speckle noise removal effect is good, the number of artificial traces is small, detail information of the image is kept, and the calculation speed is higher through the GPU.

Description

technical field [0001] The invention relates to a SAR image multiplicative coherent speckle noise removal method based on a multi-scale hole residual attention network, which belongs to the field of remote sensing image processing. Background technique [0002] Synthetic aperture radar (SAR) is a coherent imaging sensor that can acquire large amounts of high-quality surface data. Due to its ability to operate at night and in harsh weather conditions such as thin clouds and smog, SAR has gradually become an important source of remote sensing data in fields such as geographic mapping, resource investigation, and military reconnaissance. However, SAR images are often affected by multiplicative noise caused by the coherence of scattering phenomena, that is, coherent speckle noise. The existence of coherent speckle noise seriously affects the quality of SAR images, and greatly reduces the efficiency of SAR image interpretation tasks such as object detection and instance segmenta...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problem of SAR image quality degradation caused by coherent speckle noise caused by the coherence of scattering phenomena in the SAR imaging process, a high-performance and high-precision SAR image denoising algorithm is designed to obtain clean and noise-free SAR images to improve Efficiency and Accuracy of SAR Image Interpretation Task

Images