Extraction of Combined Features from Global/Local Statistics of Visual Words Using Relevant Operations

Tetsu MATSUKAWA  Takio KURITA  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E93-D   No.10   pp.2870-2874
Publication Date: 2010/10/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E93.D.2870
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Image Recognition, Computer Vision
Keyword: 
combined feature,  bag-of-features,  feature selection,  image classification,  

Full Text: PDF>>
Buy this Article




Summary: 
This paper presents a combined feature extraction method to improve the performance of bag-of-features image classification. We apply 10 relevant operations to global/local statistics of visual words. Because the pairwise combination of visual words is large, we apply feature selection methods including fisher discriminant criterion and L1-SVM. The effectiveness of the proposed method is confirmed through the experiment.