| 基于ICA的稀疏编码算法在图像处理中的应用 |
| |
| 引用本文： | 陈红艳,马上.基于ICA的稀疏编码算法在图像处理中的应用[J].成都信息工程学院学报,2013(5):481-484. |
| |
| 作者姓名： | 陈红艳 马上 |
| |
| 作者单位： | 成都信息工程学院大气探测学院;电子科技大学通信抗干扰技术国家级重点实验室 |
| |
| 基金项目： | 国家自然科学基金青年基金资助项目(61101033) |
| |
| 摘 要： | 针对独立分量分析基于数据间的高阶统计特性，并能有效揭示图像的本质特征的优势，分析了自然图像基于ICA的稀疏编码的实质，主要从图像压缩和图像去噪两个方面讨论了稀疏编码算法在图像处理中的应用。仿真实验结果表明，稀疏编码算法能够有效提取自然图像的特征基向量，利用特征系数的稀疏性，可以有效实现自然图像的压缩；结合软门限算子对噪声图像的特征系数进行处理，能有效减小自然图像中的高斯噪声的影响。
|
| 关 键 词： | 信息处理技术 图像处理 独立分量分析 稀疏编码 特征向量 |
Application of Sparse Coding Algorithm based on ICA in Image Processing |
| |
| CHEN Hong-yan;MA Shang.Application of Sparse Coding Algorithm based on ICA in Image Processing[J].Journal of Chengdu University of Information Technology,2013(5):481-484. |
| |
| Authors: | CHEN Hong-yan;MA Shang |
| |
| Institution: | CHEN Hong-yan;MA Shang;Electronic Engineering College,Chengdu University of Information Technology;National Key La boratory of Science and Technology on Communications,University of Electronics Science and Technology of China; |
| |
| Abstract: | Independent Component Analysis can reveal substantive characteristics of image because it is based on high order statistical property of data. Substance of sparse coding in natural image is analyzed with ICA. Applications of Sparse coding algorithm based on ICA in image processing are studied in two aspects of image compression and noise removal. The simulation experiments show that the method can extracts basic vector of natural images. Natural im- age is compressed effectively by using sparsity of coefficients of corresponding features. With soft-threshold operator on character coefficients in noise image, the method can greatly reduce the Gaussian Noise impaction in image. |
| |
| Keywords: | information processing technology image processing independent component analysis sparse coding eigenvector |
| 本文献已被 CNKI 维普 等数据库收录！ |
