COMPARATIVE ANALYSIS OF IMAGE DENOISING TECHNIQUES FOR ENHANCING REAL-TIME IMAGES

Enhancing the real-time images using suitable denoising technique is a challenging aspect. The main objective of this research is to find a methodology or technique that enhances the real-time images such that it gives better accuracy during future image analysis. This article proposed a method that combines the fuzzy concept from Fuzzy based Median Filter [FBMF] and Modified Harris Operator concepts from Region Classification and Response Median Filtering [RCRMF] to detect the noisy pixel of interest and represented by fuzzy flag. The mean used for replacement of processing pixel is calculated using the concept from Absolute Difference, Median and Mean filter [ADMMF]. This proposed method is used to enhance the real-time images such as Computer Tomography (CT) medical and Webcam images. The efficiency of the proposed technique is compared with Fuzzy Based Median Filter (FBMF), Median, Wiener and Bilateral Filter (MWBF) and Absolute Difference Mean and Median Filter (ADMMF) in terms of Peak Signal to Noise Ratio (PSNR), Signal to Noise Ratio (SNR), Mean Square Error (MSE), and Structural Similarity Index Measure (SSIM) metrics. The result of the experiment shows that the proposed method gives the better result than the other three existing filters.