摘要
在分析传统脊波变换去噪方法优缺点的基础上,针对其不足,提出一种基于复脊波变换的去噪方法。算法将传统脊波变换中的标量一维小波变换替换为二元树复小波变换,使得脊波变换具有平移不变性;然后,对图像采用冗余分块处理,使得处理结果更平滑,有效地提高了图像的峰值信噪比(PSNR)。仿真实验表明,在SAR图像去噪应用中,本方法能够更好地保留图像中的纹理信息,处理结果优于传统脊波变换以及小波变换去噪方法。
This paper proposed a novel image denoising method. The method incorporated the dual-tree complex wavelets into the ordinary ridgelet transform. It made the ridgelet transform get approximate shift invariant property. The new method is a very good choice for image denoising. Especially SAR image, for it needed to restore the liner detail high quality. Experimental results show that the new method is better than wavelet and the ordinary ridgelet image denoising. Complex ridgelet can be the best method for SAR image denoising.
出处
《计算机应用研究》
CSCD
北大核心
2013年第10期3149-3151,共3页
Application Research of Computers
基金
国家自然科学基金资助项目(61272286)
关键词
脊波变换
图像去噪
二元树复小波
SAR图像
ridgelet transform
image denoising
dual-tree complex wavelet
SAR image
