【摘要】：相干斑抑制是極化合成孔徑雷達(dá)(Pol SAR)圖像分析的重要預(yù)處理步驟。為了更好地抑制極化SAR圖像中的相干斑,本文綜合目標(biāo)的異質(zhì)性和結(jié)構(gòu)信息,提出基于目標(biāo)異質(zhì)性的非局部均值濾波方法。首先利用K分布距離度量目標(biāo)的異質(zhì)性,并以異質(zhì)性為基礎(chǔ),保留圖像中的點、線等高異質(zhì)性目標(biāo);然后計算圖像塊之間的異質(zhì)性差異,最后將其作為度量非局部均值加權(quán)濾波像元相似性的權(quán)重系數(shù),實現(xiàn)對Pol SAR圖像的相干斑抑制。實驗對比結(jié)果表明:本文方法能夠有效地抑制相干斑,同時對細(xì)節(jié)信息和極化信息也具有良好的保持性,能夠為后續(xù)的圖像應(yīng)用提供支持。
[Abstract]:Speckle suppression is an important preprocessing step for polarization synthetic aperture radar (SAR) (Pol SAR) image analysis. In order to suppress the speckle in polarized SAR images, a nonlocal mean filtering method based on target heterogeneity is proposed in this paper. Firstly, the K distribution distance is used to measure the heterogeneity of the target, and based on the heterogeneity, the high heterogeneity targets such as points and lines in the image are preserved. Then the heterogeneity difference between image blocks is calculated, and finally it is used as the weight coefficient to measure the similarity of non-local mean weighted filter pixels, and the speckle suppression of Pol SAR images is realized. The experimental results show that the proposed method can effectively suppress the speckle, and also has good retention to the details and polarization information, and can provide support for subsequent image applications.
【作者單位】： 中國地質(zhì)大學(xué)(武漢)信息工程學(xué)院;
【基金】：國家自然科學(xué)基金(編號:41471355,41301477)
【分類號】：TN957.52

【相似文獻】

相關(guān)期刊論文 前1條

1 陳杰;朱晶;周蔭清;李春升;;復(fù)雜目標(biāo)場景合成孔徑雷達(dá)圖像異質(zhì)性分析與測量方法研究[J];電子學(xué)報;2008年09期

，

本文編號：2494090