Mean Laplacian mappings-based difference LDA for face recognition
详细信息 查看全文
- 作者：Zhaokui Li ; Yan Wang ; Xing Zhou ; Guohui Ding…
- 关键词：Mean Laplacian mappings ; Difference scatter matrix ; Linear discriminant analysis ; Robust dissimilarity measures ; Face recognition
- 刊名：Multimedia Tools and Applications
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：76
- 期：2
- 页码：2243-2265
- 全文大小：
- 刊物类别：Computer Science
- 刊物主题：Multimedia Information Systems; Computer Communication Networks; Data Structures, Cryptology and Information Theory; Special Purpose and Application-Based Systems;
- 出版者：Springer US
- ISSN：1573-7721
- 卷排序：76
文摘
This paper proposes a difference LDA based on mean Laplacian mappings. For each pixel, we firstly estimate multiple mean Laplacian mappings which include an odd and even and full mean Laplacian mappings, and generate three different images respectively. Then, we obtain a concatenated image by concatenating the odd, even and full images. The usage of the concatenated mean Laplacian mapping results in a more robust dissimilarity measures between images. In order to derive discriminative representation for the concatenated feature vector, we introduce a difference LDA which applies a difference scatter matrix to find the subspace that best discriminates different face classes. The introduction of the difference scatter matrix avoids the singularity of the within-class scatter matrix. Experiments show that the proposed method for facial expression, illumination change and different occlusion has better robustness, and achieves a higher recognition rate. For a single sample per person, the proposed method can also obtain a higher recognition rate.