Comprehensive analysis and evaluation to unsupervised binary hashing method in image similarity measurement. Issue 8 (12th July 2017)

• Record Type: Journal Article
• Authors: Kim, Jin‐Bum; Park, Rae‐Hong; Kim, Hong‐In
• Is Part Of: IET image processing. Volume 11:Issue 8(2017)
• Page Start: 633
• Page End: 639
• Publication Date: 2017-07-12
• Subjects: data compression -- quantisation (signal) -- image coding -- cryptography -- binary codes; unsupervised binary hashing method -- image similarity measurement -- comprehensive analysis -- UBH method -- orthogonal locality preserving projection -- dimensionality reduction -- DR -- quantisation error reduction -- binary hash code -- low‐dimensional vector -- rotation optimal parameter -- offset optimal parameter -- feature space preservation -- semantic consistency; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2016.0935
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : This study proposes an unsupervised binary hashing (UBH) method and provides a comprehensive analysis and evaluation to the UBH method in image similarity. Using the orthogonal locality preserving projection, the proposed UBH method performs the dimensionality reduction (DR). To reduce the quantisation error between low‐dimensional vectors and binary hash codes generated in the DR, the proposed UBH method calculates the optimal parameters of rotation and offset. The authors use two test beds to evaluate the preservation of original feature space and the semantic consistency. Experimental results with two test beds show that the proposed UBH method is state of the art.