| Paper: | MLSP-P7.5 | ||
| Session: | Pattern Recognition and Classification II | ||
| Time: | Friday, May 21, 15:30 - 17:30 | ||
| Presentation: | Poster | ||
| Topic: | Machine Learning for Signal Processing: Signal detection, Pattern Recognition and Classification | ||
| Title: | MULTIRESOLUTION EIGENIMAGES FOR TEXTURE CLASSIFICATION | ||
| Authors: | Mehrdad Gangeh; Multimedia University | ||
| Michel Bister; University of Nottingham in Malaysia | |||
| Madasu Hanmandlu; Indian Institute of Technology, Delhi | |||
| Abstract: | Based on the Gaussian properties of eigenimages, this paper presents a new technique for texture classification using multiresolution eigenimages. The input image, composed of two textures from the Brodatz album, is subdivided into N sub-images of fixed size d´d, which are blurred with a Gaussian and normalized. The application of Hotelling transform decomposes each sub-image into d^2 eigenimages. The R largest resulting coefficients can be used for classification of the texture present in the sub-images. Classification is done using the fuzzy C-means (FCM) algorithm and the performance is measured with an appropriate quality factor. We discuss the successful application of this technique, as well as the influence of the different parameters of the classification process on several pairs of textures. Moreover, combination of Hotelling coefficients obtained with different values of d is shown to improve the performance, based on the idea of analyzing the texture at different levels of resolution. | ||
| Back | |||
Home -||-
Organizing Committee -||-
Technical Committee -||-
Technical Program -||-
Plenaries
Paper Submission -||-
Special Sessions -||-
ITT -||-
Paper Review -||-
Exhibits -||-
Tutorials
Information -||-
Registration -||-
Travel Insurance -||-
Housing -||-
Workshops