Multi-stage Detection based Filtering for Removal of Impulse Noise from Images

Abstract

This paper presents a new technique for removal of impulse noise from images. By incorporating the rank-conditioned median filtering, a threshold operation on the deviation of the extreme ranks from the healthy pixel-values in the filtering window and the center-weighted median (CWM) based adaptive impulse detection mechanism into the detection framework. An improved impulse detection scheme is developed. The proposed detection scheme followed by rank-ordered mean (ROM) filtering of corrupted pixels effectively reduces impulse noise while preserving the image details. Simulations show that the proposed scheme works better than the existing schemes in suppressing both fixed-valued and random-valued impulses at different noise ratios.

Indexing terms:

• Fixed-valued impulse noise
• Random-valued impulse noise
• Rank-conditioned median filter
• Center-weighted median filter
• Rank-ordered mean and threshold

Additional information

Notes on contributors

Kh Manglem Singh

Kh Manglem Singh born in 1963, did his BE (Electrical), ME (Electrical) and MS (Software Systems) from Dayalbagh Educational Institute, Agra, Delhi College of Engineering, Delhi and Birla Institute of Technology and Science, Pilani respectively. He is currently a research scholar at Indian Institute of Technology Guwahati. His areas of interest are digital image processing, digital signal processing, fuzzy logic and its applications and data security

Prabin K Bora

P K Bora born on 1957, did his BE from Assam Engineering College, Guwahati and ME and PhD from Indian Institute of Science, Bangalore all in Electrical Engineering. He is currently an associate Professor in the Department of Electronics and Communication Engineering, Indian Institute of Technology Guwahati. His research interest includes image processing and computer vision