Improved fuzzy image enhancement using L*a*b* Color space and edge preservation

Abstract

Image enhancement is a process of improving the perceptibility of an image so that the output image is better than input image. The traditional image enhancement techniques may affect the edges of an image which leads to loss of perceptual information. The existing techniques use primary/secondary color spaces which are device-dependent. This research paper works on these two issues. It uses L*a*b* color space which is device independent. To evaluate fuzzy membership values, L component is stretched while preserving the chromatic information a and b. Moreover, an edge preserving smoothing has been integrated with fuzzy image enhancement so that edges are not affected and remain preserved. The proposed technique is compared with existing techniques such as Histogram equalization, Adaptive histogram equalization and fuzzy based enhancement. The experimental results indicate that the proposed technique outperforms the existing techniques.