摘要
Abstract
Supervised learning is a common learning approach in pattern classification, which needs to use samples with labels to adjust the parameters of the classifier in order to achieve the correct classification. But in practical application, it is limited, even not easy to obtain labeled training samples. Therefore, the unsupervised learning that only needs samples without labels gradually becomes one of the hot research topics. Based on Hebb learning theory and principal component analysis, the paper adopts the unsupervised learning method to realize the compression of digital images. The experimental results indicate that the algorithm can well realize the compres-sion of digital images and has certain actual application value.关键词
Hebb理论/机器学习/无监督学习/主分量分析/图像压缩Key words
Theory of Hebb/Machine Learning/Unsupervised Learning/PCA/Image Compression分类
信息技术与安全科学
