Texture classification by multifractal spectrum and barycentric coordinates of bit planes of wavelet coefficients. Issue 12 (17th October 2017)

• Record Type: Journal Article
• Authors: Wang, Shuli; Wang, Guanxiang
• Is Part Of: IET image processing. Volume 11:Issue 12(2017)
• Page Start: 1205
• Page End: 1209
• Publication Date: 2017-10-17
• Subjects: image texture -- image classification -- spectral analysis -- wavelet transforms -- fractals -- image reconstruction -- singular value decomposition; UMD database -- Brodatz database -- weighted L1 distance -- 1NN classifier -- one‐nearest‐neighbour classifier -- SVD reconstruction -- SVD decomposition -- low‐frequency domain -- three‐level high‐frequency domains -- fractal dimensions -- FDBC -- signature vector -- wavelet transform -- wavelet coefficients -- bit planes -- barycentric coordinates -- multifractal spectrum -- texture classification; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2016.0875
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : A new texture classification method based on wavelet transform is presented. The elements of the signature vector, FDBC, of an image are the fractal dimensions and barycentric coordinates of the bit planes of the wavelet coefficients in both the three‐level high‐frequency domains and the third low‐frequency domain. The pretreatment is done with SVD decomposition and reconstruction by dropping half singular values. The one‐nearest‐neighbour classifier (1NN) with L 1 distance is used to make the classification. Furthermore, to improve classification result, the classifier 1NN is strengthened with weighted L 1 distance. The proposed method is tested on five subsets from Brodatz database and UMD database and is experimentally proved more efficient and more promising.