A Survey on various Image Filtering Approaches to remove Impulse Noise

Abstract: Images gets corrupted either during acquisition or transmission. Frequently occurring noise that might occur in images is impulse noise, because of that various image processing operations such as image segmentation, object identification, and similarity matching etc. cannot be performed efficiently. This paper focus on various existing image filtering techniques and their improvements. Several median-based denoising methods tends to work well for low level impulse noise but perform poorly for high level impulse noise. Index Terms- Noise models, threshold based switching median filter, operator based median filter, morphological based median filter, statistics based median filter.