Dynamic Histogram Equalization for contrast enhancement for digital images

Abstract

Magnetic Resonance Imaging (MRI) is an efficient tool, produced by applying radio waves and magnetic fields which is being useful in the diagnosis of various diseases like cancer, epilepsy and stroke etc. The quality of the resulting image is needed to be enhanced because it is challenging for the specialists to investigate. Modified Histogram Equalization on Fuzzy based Improved Particle Swarm Optimization (FIPSO) is proposed for Dynamic Histogram Equalization which resolves this problem through image contrast enhancement. The details of an images are captured by smoothing and it uses Gaussian function to distribute pixel intensity to nearest pixel. It uses normal distribution and here blur is removed by applying Non subsampled Contourlet Transform. Then local maxima are calculated to extract dark and bright pixel values. The smoothed images are fuzzified with TSK (Takagi-Sugeno-Kang) model and it provides importance to all the local maxima intervals. An Improved particle swarm optimization (IPSO) algorithm is obtained by combining Galactic Swarm Optimization (GSO) with PSO which equalizes histogram of an image. FIPSO algorithm is used to the minimum contrast images of MRI brain images. Non-subsampled Contourlet transform (NSCT) based modified histogram equalization enhances image contrast. Here IPSO generates optimum values and these value are used to calculate cumulative distribution function in histogram equalization. The quality measures demonstrate that the current equalization technique attains highest performance against existing techniques in terms of brightness and contrast.