Pattern Spectra for Texture Segmentation of Gray-Scale Images

Velloso, M.L.F.; Carneiro, Thales Ávila; Souza, F.J. de

doi:10.1109/isda.2007.150

Cited by 3 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experiments showed that the most non-Gaussian components of each analyzed texture were able to cluster the test samples. Velloso et al 12 presented an unsupervised segmentation of textured images that combines local pattern spectra features and dimensionality reduction techniques by PCA and ICA. They used two neural PCA and ICA algorithms on the features extracted for achieving redundancy reduction and noise reduction.…”

Section: Related Workmentioning

confidence: 99%

“…ICA has been proposed as a generic statistical model for images and applied to texture analysis. [4][5][6][7][8][9][10][11][12][13][14][15] This paper presents a supervised method for texture classification by ICA of Gabor features ͑called ICAG͒. First, training samples are randomly selected from the texture image, and a Gabor wavelet family is defined by referring to the design strategy proposed by Jain and Farrokhnia, 16 which is able to avoid analyzing the Fourier spectrums of the texture.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Independent component analysis of Gabor features for texture classification

Chen

Han

et al. 2008

Opt. Eng

View full text Add to dashboard Cite

Extracting effective features for texture classification is always a difficult problem in texture analysis. We present a method for texture features extraction by independent component analysis ͑ICA͒ of Gabor features ͑called ICAG͒. It has three distinguished aspects. First, Gabor wavelet transformation first produces distinct textural features characterized by spatial locality, scale, and orientation selectivity. Second, principal component analysis ͑PCA͒ and ICA then reduce the dimensionality and redundancy of these features. Thus, the resulting independent components are taken as texture features for classification. Third, in the ICA procedure, two different frameworks are discussed. Framework I, called ICAG I, regards pixels as random variables and represents them as a column vector by reshaping all the transformed images row by row, while framework II, called ICAG II, treats the statistical features, namely, the mean and standard deviation, as random variables. Thus, the statistical features of all the transformed images construct a column vector. Finally, comparative experiments between ICAG and the other traditional methods-ICA and Gabor wavelets-are performed on two composite images ͑five narrowband textures and five natural textures with broader bands͒. The results indicate that ICAG provides the best performance and ICAG II is more efficient and applicable.

show abstract

Section: Related Workmentioning

confidence: 99%