Genetic-Based Multiresolution Noisy Color Image Segmentation
Segmentation of a color image composed of different kinds of regions can be a hard problem, namely to compute for an exact texture fields and make a decision of the optimum number of segmentation areas in an image when it contains similar and/or unstationary texture fields. A local novel neighborhood-based segmentation approach is proposed. Genetic algorithm is used in the proposed limited segment-pass optimization process. In this pass, an energy function, which is defined based on Markov Random Fields, is minimized. The proposed system uses an adaptive threshold estimation method for image thresholding in the wavelet domain based on the Generalized Gaussian Distribution (GGD) modeling of sub band coefficients. This method called Normal Shrink is computationally more efficient and adaptive because the parameters required for estimating the threshold depend on sub band data energy that used in the pre-stage of segmentation. A quadtree is utilized to implement the fast clustering segments for multiresolution framework analysis, which enables the use of different strategies at different resolution levels, and hence, the omputation can be accelerated. The experimental results of the proposed segmentation approach are very encouraging
- Each author retains the right to use the work for non-commercial purposes as well as for further research and spoken presentations.
- Each author retains the right to use the illustrations and research data in his/her future work.
- Only one offprint is provided free for each author. The authors can order offprints at the proof stage at certain rates depending on the number of additional copies required and the year of publication.
The publisher of the journal has full rights for publication of the submitted manuscripts, electronic and facsimile formats and for electronic capture, reproduction and licensing in all formats now and in perpetuity in the original and all derivative works.