摘要
提出一种基于小波系数和最小二乘支持向量机(LS-SVM)的图像去噪方法。根据小波系数的性质,依据邻域小波系数的平均值来选取特征向量来进行训练,然后用训练得到的LS-SVM分类器将含噪图像中的像素分为噪声或非噪声点,进行去噪处理。实验结果表明该方法能达到较高的峰值信噪比,具有很好的去噪效果。
The least squares support vector machine （LS-SVM） is a modified version of SVM, which simplifies the complexity of optimization problem of the SVM algorithm. In this paper, a wavelet-based image denoising using LS-SVM is proposed. According to the nature of the wavelet coefficients, the average of the neighborhood of wavelet coefficients as the feature vectors for training are selected. Then the noisy image pixels are divided into noise and non-noise pixels by the training LS-SVM classifier and noise processing. The experimental results show that, by using this method, a higher PSNR could be achieved,and a better donoising effective could be have.
出处
《电视技术》
北大核心
2013年第1期18-20,23,共4页
Video Engineering
基金
国防预研基金资助项目(513060602)
