Journal
JOURNAL OF MODERN OPTICS
Volume 70, Issue 8, Pages 503-516Publisher
TAYLOR & FRANCIS LTD
DOI: 10.1080/09500340.2023.2273552
Keywords
Feature extraction; similarity sequence; correlation analysis; image recognition
Categories
Ask authors/readers for more resources
This paper proposes a novel multi-view discriminant feature extraction method that considers the similarity sequence between samples. It constructs similarity-sequenced discriminant scatters to preserve the sequence structure of within-class samples and develops between-class correlations to constrain the intrinsic manifold structure of cross-view samples. Experimental results show that the proposed method achieves higher recognition accuracy and stronger robustness in image recognition tasks.
Traditional multi-view feature extraction methods based on manifold learning frequently overlook the similarity sequence between samples, failing to capture the intrinsic manifold structure of raw nonlinear samples and restricting the recognition performance of multi-view learning. In this paper, we propose a novel similarity-sequenced multi-view discriminant feature extraction method, called Similarity -sequenced Multi-view Discriminant Correlation Analysis (SMDCA), which explicitly considers the sample sequences based on similarity. The method constructs similarity-sequenced discriminant scatters for preserving the sequence structure of within-class samples and develops between-class correlations with the similarity-sequence structure information for further constraining intrinsic manifold structure of cross-view samples. SMDCA can also simultaneously extract low-dimensional sequence features with well-discriminative power from multiple views. Extensive experiments exhibit that SMDCA can provide higher recognition accuracy and stronger robustness in image recognition tasks.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available