Open AccessSoft Computing Tool Approach for Texture Classification Using Discrete Cosine Transform
Author(s) -
Pankaj H. Chandankhede
Publication year - 2011
Publication title -
international journal of electronic signal and systems
Language(s) - English
Resource type - Journals
ISSN - 2231-5969
DOI - 10.47893/ijess.2011.1003
Subject(s) - discrete cosine transform , soft computing , artificial intelligence , computer science , artificial neural network , backpropagation , pattern recognition (psychology) , pixel , image texture , fuzzy logic , feedforward neural network , texture (cosmology) , computer vision , image processing , image (mathematics)
Texture can be considered as a repeating pattern of local variation of pixel intensities. Cosine Transform (DCT) coefficients of texture images. As DCT works on gray level images, the color scheme of each image is transformed into gray levels. For classifying the images using DCT, two popular soft computing techniques namely neurocomputing and neuro-fuzzy computing are used. A feedforward neural network is used to train the backpropagation learning algorithm and an evolving fuzzy neural network to classify the textures. The soft computing models were trained using 80% of the texture data and the remaining was used for testing and validation purposes. A performance comparison was made among the soft computing models for the texture classification problem. In texture classification the goal is to assign an unknown sample image to a set of known texture classes. It is observed that the proposed neuro-fuzzy model performed better than the neural network.