High probability impulsive noise removal for corrupted images

Ju Yuan Hsiao; Shu Yi Chen

High probability impulsive noise removal for corrupted images

Department of Computer Science and Information Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We know one characteristic of image is that each pixel is very close to surrounding neighbors. If the impulsive noise appears in the local area, it will be very abruptly. In the proposed scheme, we pick up several neighboring pixels around the center pixel in the scanning window as the statistic factors. The mean value and mean of absolute neighboring difference (MAND) are utilized from those. Further, we use them to acquire a critical value. The value is a representative characteristic of those pixels in the local area and it will be used to identify whether if a pixel is corrupted.

Original language	English
Title of host publication	Proceedings - 24th International Conference on Distributed Computing Systems Workshops
Editors	N.-F. Tzeng, M. Takizawa
Pages	190-194
Number of pages	5
Volume	24
Publication status	Published - 2004 Jul 12
Event	Proceedings - 24th International Conference on Distributed Computing Systems Workshops - Hachioji, Japan Duration: 2004 Mar 23 → 2004 Mar 24

Other

Other	Proceedings - 24th International Conference on Distributed Computing Systems Workshops
Country	Japan
City	Hachioji
Period	04-03-23 → 04-03-24

All Science Journal Classification (ASJC) codes

Software
Hardware and Architecture
Computer Networks and Communications

Access to Document

Fingerprint Dive into the research topics of 'High probability impulsive noise removal for corrupted images'. Together they form a unique fingerprint.

Cite this

@inproceedings{6667fa9323254f4d8f6b71d25090973b,

title = "High probability impulsive noise removal for corrupted images",

abstract = "We know one characteristic of image is that each pixel is very close to surrounding neighbors. If the impulsive noise appears in the local area, it will be very abruptly. In the proposed scheme, we pick up several neighboring pixels around the center pixel in the scanning window as the statistic factors. The mean value and mean of absolute neighboring difference (MAND) are utilized from those. Further, we use them to acquire a critical value. The value is a representative characteristic of those pixels in the local area and it will be used to identify whether if a pixel is corrupted.",

author = "Hsiao, {Ju Yuan} and Chen, {Shu Yi}",

year = "2004",

month = jul,

day = "12",

language = "English",

volume = "24",

pages = "190--194",

editor = "N.-F. Tzeng and M. Takizawa",

booktitle = "Proceedings - 24th International Conference on Distributed Computing Systems Workshops",

note = "Proceedings - 24th International Conference on Distributed Computing Systems Workshops ; Conference date: 23-03-2004 Through 24-03-2004",

}

TY - GEN

T1 - High probability impulsive noise removal for corrupted images

AU - Hsiao, Ju Yuan

AU - Chen, Shu Yi

PY - 2004/7/12

Y1 - 2004/7/12

N2 - We know one characteristic of image is that each pixel is very close to surrounding neighbors. If the impulsive noise appears in the local area, it will be very abruptly. In the proposed scheme, we pick up several neighboring pixels around the center pixel in the scanning window as the statistic factors. The mean value and mean of absolute neighboring difference (MAND) are utilized from those. Further, we use them to acquire a critical value. The value is a representative characteristic of those pixels in the local area and it will be used to identify whether if a pixel is corrupted.

AB - We know one characteristic of image is that each pixel is very close to surrounding neighbors. If the impulsive noise appears in the local area, it will be very abruptly. In the proposed scheme, we pick up several neighboring pixels around the center pixel in the scanning window as the statistic factors. The mean value and mean of absolute neighboring difference (MAND) are utilized from those. Further, we use them to acquire a critical value. The value is a representative characteristic of those pixels in the local area and it will be used to identify whether if a pixel is corrupted.

UR - http://www.scopus.com/inward/record.url?scp=3042590263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3042590263&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:3042590263

VL - 24

SP - 190

EP - 194

BT - Proceedings - 24th International Conference on Distributed Computing Systems Workshops

A2 - Tzeng, N.-F.

A2 - Takizawa, M.

T2 - Proceedings - 24th International Conference on Distributed Computing Systems Workshops

Y2 - 23 March 2004 through 24 March 2004

ER -