基于LS-SVM的复轮廓波变换的图像去噪

针对传统的轮廓波变换图像去噪时引入边缘混叠现象,提出了复轮廓波变换(Complex Contourlet Transform,CCT)和最小二乘支持向量机(LS-SVM)的图像去噪方法.该方法充分利用了复轮廓变换的平移不变性、多方向性以及LS-SVM的小样本学习能力,应用训练好的LS-SVM模型将含噪图像的CCT系数分为含噪点和非含噪点,进行去噪处理.仿真结果表明该算法有效保护图像边缘纹理信息,其峰值信噪比明显高于其他算法,并且具有良好的视觉效果.

Image Denoising Using LS-SVM classification in complex contourlet transform domain

Concerning edge aliasing phenomenon is introduced by traditional contourlet transform in image denoising,a new denosing method which incorporate the least squares support vector machine into the complex contourlet transform is proposed in this paper.This method makes full use of the translational invariance and multi-directional feature of complex contourlet conversion, and learning ability of small samples in LS-SVM, the noisy image CCT coefficients are divided into noise and non-noise pixels by the training LS-SVM classifier and noise processing.Simulation results show that the algorithm can effectively protect the image edge and texture information,the peak signal to noise ratio significantly better than the other algorithms,and a good visual effect is presented.