Open AccessVector Isolated Minimum Distance Filtering for Image De-Noising in Digital Color Images
Author(s) -
Praveen B. Choppala,
James Stephen Meka,
P. Sreenivasula Reddy
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.d7174.118419
Subject(s) - artificial intelligence , pixel , smoothing , median filter , computer vision , color image , computer science , digital image , component (thermodynamics) , image (mathematics) , pattern recognition (psychology) , image processing , mathematics , physics , thermodynamics
Image de-noising forms a crucial component of digital image processing. The state-of-the-art vector median filtering based image de-noising approaches like the median filtering, the vector median filtering and the basic vector directional filtering and their extensions process the vector pixels jointly in the red, green and blue components. Consequently any smoothing applied therein is leveraged on all the color components equally. In this paper we propose that processing the vectors in isolation, that is, each color component taken separately, and then smoothed by minimising the aggregate distance between the pixels in each color component will lead to more efficient de-noising of noisy images. We demonstrate the superiority of the proposed method compared against vector filtering approaches using several images and test measures.