小波支持向量去噪

Laser speckle image de-noising based on wavelet support vector machine

噪声对图像质量产生不利影响,降低了图像的应用价值,为了更好的消除图像中的噪声,提出一种小波支持向量去噪方法。首先采集大量的激光散斑图像,并将它们划分成为训练图像集和测试图像集,然后将训练图像集输入到小波支持向量机进行学习,将图像中信号划分为噪声和有用信息,最后采用测试图像集对小波支持向量机去噪性能进行测试与分析。实验结果表明,相比于其它激光散斑图像去噪方法,小波支持向量机可以更加有效的去除噪声,可以很好的保护图像边缘细节信息,提高了激光散斑图像的质量。

Noise has some adversely affect image on quality and reduces the application value of the image,in order to eliminate the noise in the image,this paper proposed a laser speckle image de-noising method based on wavelet support vector machine.Firstly,large numbers of laser speckle images are collected and are divided into the set of training images and testing images set,and then a training image set are input to wavelet support vector machine to learn and the establish classify model for noise and useful information of the image,finally,test image set are used to test the de-noise performance the wavelet support vector machine.The experimental results show that,compared with other laser speckle image de-noising methods,the wavelet support vector machine can remove noise more effective,can protect edge detail information and improve the quality of laser speckle image.