A pixel value restoration considerng line stucture for impulse noise removal

In this paper, a pixel value restoration method for images corrupted by impulse noise is proposed. For example, the median filter is a well-known solution for impulse noise removal. However, it cannot process well at line structures in images. We therefore propose an effective method in the view point of line structure restoration. The proposed method considers line structures in four angles. The key point of the method is to consider the two adjacent lines of the line which pass through the target pixel.

[1]  Yoshinobu Kajikawa,et al.  Directional difference filter: its effectiveness in the postprocessing of noise detectors , 2002 .

[2]  Sung-Jea Ko,et al.  Center weighted median filters and their applications to image enhancement , 1991 .

[3]  Dehua Li,et al.  A switching vector median filter based on the CIELAB color space for color image restoration , 2007, Signal Process..

[4]  Yrjö Neuvo,et al.  Detail-preserving median based filters in image processing , 1994, Pattern Recognit. Lett..

[5]  Kai-Kuang Ma,et al.  A switching median filter with boundary discriminative noise detection for extremely corrupted images , 2006, IEEE Trans. Image Process..

[6]  Noriaki Suetake,et al.  Properties and Effective Extensions of Local Similarity-Based Pixel Value Restoration for Impulse Noise Removal , 2012, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[7]  Jinsung Oh,et al.  Minimum-Maximum Exclusive Interpolation Filter for Image Denoising , 2007, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[8]  Akira Taguchi,et al.  Removal of impulse noise from highly corrupted images by using noise position information and directional information of image , 2001, IS&T/SPIE Electronic Imaging.

[9]  Yoshinobu Kajikawa,et al.  Directional difference‐based switching median filters , 2002 .

[10]  Suetake Noriaki,et al.  Impulse Noise Removal Based on Directional Variance for Text Images , 2010 .

[11]  John W. Tukey,et al.  Nonlinear (nonsuperposable) methods for smoothing data , 1974 .