A genetic algorithm-based segmentation of Markov random field modeled images

Colour image segmentation with histogram and homogeneity histogram difference using evolutionary algorithms

“…To evaluate the quality of segmented image, a method has described by Kim et al [16], is considered. Here, the evaluation function F q is defined as.…”

“…In the past few decades, metaheuristics have emerged as a significant tool for segmentation of colour images out of the various metaheuristics, GA [19] are probably the most frequently used technique for image segmentation [1,[8][9][10][11][12][13][14][15][16]. Some other metaheuristics used for image processing are: PSO [17,3], DE [18], ABC [3] [43], Cuckoo Search [24] etc.…”

Genetic algorithms for video segmentation

“…The purpose of DGAs is to improve the quality of the final solution and accelerate the convergence. Studies have shown that DGAs can be generalized to an unsupervised segmentation problem and are effective for segmenting images [7,8]. Furthermore, Park et al proposed a DGA-based segmentation method for traffic image sequences in Ref.…”

“…This algorithm is helpful for improving the search efficiency, while alleviating local minima problems associated in deterministic optimization algorithms. In contrast to hybrid methods using other optimization algorithms, distributed genetic algorithms (DGAs) have been used to segment gray-level images, textured images, and color images [7,8]. DGAs constrain the interaction of each chromosome to a limited number of chromosomes, rather than the whole population.…”

“…Among these methods, there has been considerable interest in the Markov random field (MRF)-based segmentation method [4][5][6][7][8][9][10][11]. An MRF is effective in representing the spatial and temporal continuity between adjacent pixels in various applications, and robust to degradation.…”