Grayscale Image Enhancement Using Water Cycle Algorithm

Recent developments in engineering and computer sciences have heightened the need for digital image enhancement. Most of the previously reported works, however, focused on image enhancement using classical methods like mathematical transformations and spatial and frequency-domain methods. Hence, recently, there has been an increasing interest in using nature-inspired optimization techniques for image processing purposes. The water cycle algorithm (WCA) is one of the nature-inspired algorithms (NIAs) that have gotten much attention in optimizing real-world engineering problems due to its appealing performance. However, to the best of the author’s knowledge, little research has been undertaken on the WCA’s image-enhancing capacity. Thus, this work is intended to offer a modified histogram equalization (HE) approach using WCA to enhance the contrast of an image and maintain its brightness. Besides, the proposed WCA-based image enhancement technique was compared to linear contrast stretching (LCS), HE and its modified versions, particle swarm optimization (PSO), and accelerated particle swarm optimization (APSO). In addition to the objective function fitness, 11 full reference (FR) image quality assessment (IQA) metrics were employed to evaluate image quality and compare performance. Experimental results showed that the suggested image enhancement technique exhibited better performance than others in enhancing dark grayscale images in terms of objective function fitness and perceptual visual IQA metrics like multi-scale structural similarity (MS-SSIM), information-weighted structural similarity (IW-SSIM), information-weighted mean squared error (IW-MSE), and information-weighted peak signal-to-noise ratio (IW-PSNR). The proposed method also demonstrated a faster convergence time to an optimum solution.