计算机工程与应用2019，Vol.55Issue(6)：197-203,7.DOI:10.3778/j.issn.1002-8331.1711-0434
图正则化迁移稀疏概念编码的跨域图像分类
Cross-Domain Image Classification with Graph Regularization Transfer Sparse Concept Coding
摘要
Abstract
In order to overcome the difference of features between different image domains and the gap of distribution, a learning algorithm based on co-regularized sparse concept encoding is proposed in this paper. Firstly, the distribution dif-ference and the label consistency information of image datasets are incorporated into the sparse coding model to study the robust sparse representation of the cross-domain image. Then, the low-dimensional manifold structure is excavated from the high-dimentional image feature space to form vector set, which contructs transfer sparse coding for robust image repre-sentation. The method captures the commonality underlying of the different image dataset and realizes cross-domain trans-fer for image tags. The experiment shows that the method achieves more robust feature representation, and its classifica-tion performance is significantly better than other related methods.关键词
稀疏编码/流形结构/基学习/标签相关性/共同特征表达Key words
sparse coding/manifold structure/basis learning/label relevance/common feature expression分类
信息技术与安全科学引用本文复制引用
孙登第,孟欠欠,马云鹏..图正则化迁移稀疏概念编码的跨域图像分类[J].计算机工程与应用,2019,55(6):197-203,7.基金项目
国家自然科学基金(No.61203056) （No.61203056）
江苏省普通高校研究生创新计划资助项目(No.CXLX11_0198). （No.CXLX11_0198）
