Optimization of Image Enhancement using an Artificial Immune System
Publication Type:Journal Article
Source:International Journal of Mechanical Engineering and Technology, Volume 8, Issue 8, p.184-192 (2017)
Keywords:Department of Science and Humanities
Using fuzzy logic technique, a new method is proposed for the enhancement of images. The suggested method is introduced for the improvement of color images using a combination of fuzzy logic method and artificial immune system optimization algorithm. The fuzzification of underexposed region has been carried out by Gaussian membership function and for fuzzifying overexposed areas, triangular membership function is found suitable. An estimate called exposure, which serves as a divider has been defined for both underexposed and overexposed parts of the image. The hue, saturation and intensity (HSV) color space is used for the transformation of the image from RGB to HSV space without changing the true color composition unblemished. Two functions, namely parametric sigmoid function and power-law operator finds good suitability for the enhancement of the underexposed and overexposed regions. An objective measure involving entropy, the contrast and the visual factors are constructed for the computation of the parameters present in the operators and are optimized using artificial immune system algorithm. The results are compared with genetic algorithm (GA) based image enhancement and the proposed method when correlated with recent GA- based approach is found to be generating better results.