Performance Analysis of Block PSO for Image De-noising using Wavelet Transform

Image de-noising is one of the fundamental problems in the field of image processing needed for improving the image quality before performing different high-level vision tasks. Numerous wavelet based denoising methods were utilized for performing image de-noising process. In such works, there is a lack of analysis in selecting the appropriate threshold value. Moreover, such analysis leads to the determination of static threshold value. The basic formulae exist if we treat noisy image as a single image without dividing it into blocks. We can also check the performance of the conventional methods by dividing the noisy image into different block sizes and then applying dynamic methods to choose proper threshold value. In this paper, we proposed an adaptive image de-noising technique by dividing the noisy image into blocks then applying wavelet transform on it and then by applying Particle Swarm Optimization (PSO) technique to select proper threshold values. The performance of the image de-noising technique is evaluated by comparing the result of proposed technique with the conventional soft thresholding technique in terms of peak signal-to-noise ratio (PSNR).