Hybrid Evolutionary Technique for Contrast Enhancement of Color Images

Abstract

AbstractImage contrast enhancement is an important task in image processing, which intends to restore the true image from degraded images. The images may be degraded due to noise or blur, and image contrast enhancement performs tasks like deblurring, denoising, etc. Digital images are very essential and important part of our everyday life. The enhancement of contrast is an effective step in image processing as well as in computer vision applications. The histogram equalization technique is used to enhance the contrast of the image, but the histogram equalization technique is not popular due to the artificial effect, immense brightness change, and over enhancement. In this paper, we have proposed a hybrid technique for image contrast enhancement based on an evolutionary algorithm. The proposed technique has been applied to many low-contrast images, and the performance has been compared with the Cuckoo search technique, artificial bee colony technique, and other similar techniques. The nonlinear models are well-optimized with the aid of bio-inspired optimization algorithms to operate adaptively with the noise and blurring models. The proposed approach has shown adequate image enhancement with average results of 26.11 as PSNR score, 620.43 MSE, 24.85 RMSE, 0.5943 UQI, 205.44 MAE, 0.4445 NAE, 4.2992 entropy, and 3.7053 mutual information. Hence, the developed algorithm will be a good compromise to operate between contrast enhancement and denoising techniques.KeywordsEvolutionary techniqueContrast enhancementImage processingDenoising