基于非下采样剪切波变换与引导滤波结合的遥感图像增强

发布时间：2018-05-25 10:05

本文选题：遥感图像 + 图像增强　；参考：《计算机应用》2016年10期

【摘要】：针对遥感图像中对比度低、细节信息缺失和边缘梯度保持能力较弱等问题,提出了一种基于非下采样剪切波变换(NSST)与引导滤波相结合的遥感图像增强算法。首先,原始图像通过NSST被分解成低频子带和高频子带两部分。然后,对低频子带进行线性增强,提高整体对比度;采用自适应阈值法抑制高频子带的噪声,再对去噪后的高频子带进行引导滤波增强,提高图像的细节信息和边缘梯度保持能力。最后,对两部分子带进行NSST反变换,得到增强后的图像。实验结果表明,与直方图均衡、基于Contourlet变换和模糊理论的图像增强算法、基于非下采样Contourlet变换与反锐化掩膜结合的遥感图像增强算法以及基于非下采样Shearlet变换与参数化对数图像处理相结合的遥感图像增强算法相比,该算法的图像信息熵、峰值信噪比(PSNR)和结构相似性(SSIM)都有一定的提升,能明显地改善图像视觉效果,使得图像纹理更加清晰。
[Abstract]:Aiming at the problems of low contrast, lack of detail information and weak edge gradient retention in remote sensing images, a remote sensing image enhancement algorithm based on non-downsampling shear wave transform (NSST) and guided filtering is proposed. First, the original image is decomposed into low frequency subband and high frequency subband by NSST. Then, the low frequency subband is linearly enhanced to improve the overall contrast. The adaptive threshold method is used to suppress the noise of the high frequency subband, and then the denoised high frequency subband is enhanced by guided filtering. Improve the image's detail information and edge gradient retention ability. Finally, the two subbands are inversely transformed by NSST, and the enhanced images are obtained. The experimental results show that the image enhancement algorithm based on Contourlet transform and fuzzy theory is balanced with histogram. Compared with the remote sensing image enhancement algorithm based on the combination of non-downsampling Contourlet transform and de-sharpening mask, and the remote sensing image enhancement algorithm based on non-downsampling Shearlet transform and parameterized logarithmic image processing, the image information entropy of this algorithm is compared with that of non-downsampling Shearlet transform and parameterized logarithmic image processing. Both PSNR (PSNR) and structural similarity (SSIM) can improve the visual effect of the image and make the texture of the image more clear.
【作者单位】：新疆大学信息科学与工程学院;上海交通大学图像处理与模式识别研究所;奥克兰理工大学知识工程与发现研究所;
【基金】：教育部促进与美大地区科研合作与高层次人才培养项目(DICE2014-2029)~~
【分类号】：TP751

【相似文献】