Improved Filtering of Noisy Images by Combining Average Filter with Bacterial Foraging Optimization Technique

Abstract

Biologically inspired algorithms have attracted a large number of researchers and found to have application in many areas of computer science including image processing. This paper presents the application of a biologically inspired algorithm, bacterial foraging optimization algorithm (BFOA), to optimize the filtering of images for denoising and its performance comparison with existing noise reduction techniques like average filter. Usually, images suffer from noises which will corrupt image quality and appearance. Hence, denoising of images plays a great role in the image processing and the frequently used filters are average filter, median filter, etc., to remove noises. This research paper explores the suitability of applying BFOA on the filtered image produced by average filter to result a further denoised image. In this proposed method of bacterial foraging-based optimization, peak signal-to-noise ratio (PSNR) is used as fitness function to denoise the noisy images. The implemented code is tested for noisy images (Gaussian noise and salt–pepper noise) filtered with average filter, and results show the optimization capability of BFOA-based method and that it improves the denoised images produced by average filter.