A Speckle Removal Method for SAR Image in Wavelet Domain Based on Heterogeneity Pre-correction

Abstract

A wavelet-domain SAR image speckle removal method based on heterogeneity pre-correction. In the wavelet domain, by counting the local heterogeneity of the SAR image, the heterogeneity information is combined with the pre-correction function to smooth the homogeneity before image speckle removal. region, protect the texture information of heterogeneous regions, and improve the problem of over-smoothing in the process of image speckle removal. This method has good speckle removal effect, low computational complexity of the pre-correction algorithm, and simple process, which solves the problems existing in the existing SAR image speckle removal methods. The processing of image signal is weakly targeted, insensitive to noise, weak in anti-interference ability, unable to reflect the subtle differences between sub-blocks in detail, poor image speckle removal effect, complex calculation, cumbersome process and other problems.

Description

Technical field: [0001] The invention relates to a wavelet domain SAR image speckle removal method based on heterogeneity pre-correction, and belongs to the technical field of SAR image processing. Background technique: [0002] Synthetic aperture radar (SAR) is regarded as a new generation of remote sensing information source because of its all-time, all-weather, multi-polarization, multi-angle data acquisition capabilities and good penetration performance. However, since the SAR system uses coherent waves for imaging, the coherent speckle noise seriously reduces the interpretability of SAR images and affects the applications of target detection, classification, recognition and information extraction. Therefore, it is necessary to study the method of coherent speckle suppression in SAR images. [0003] With the application of wavelet tools in SAR image processing, wavelet-based speckle removal methods have become an important branch of SAR image speckle removal methods. Am...

AI Technical Summary

Problems solved by technology

In the wavelet domain, by counting the local heterogeneity of the SAR image, the heterogeneity information is combined with the pre-correction function to smooth the homogeneous area before image despeckling, protect the texture information of the heterogeneous area, and improve the image over-smoothing in the process of despeckling The problem of this method is that the speckle removal effect is good, the calculation complexity of the pre-correction algorithm is low, and the process is simple, which solves the problem that the SAR image speckle removal method in the prior art has weak pertinence for image signal processing, is not sensitive to noise, and has good anti-interference ability. Weak, it cannot reflect the subtle differences between sub-blocks in detail, the effect of image despotting is not good, the calculation is complicated, the process is cumbersome, etc.

Images