Presentation 2008-03-10
Extraction of Object Regions Using Shape Map and Local Features
Koji IMAEDA, Miyako BABA, Nobuyuki SHIRAKI, Akihiro WATANABE, Yasushi HIRANO, Shoji KAJITA, Kenji MASE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, a novel method for extraction of a precise object region from a given rough initial region is proposed. The existing method for object region extraction using a kernel density estimator has problems, e.g. influence by the background edge in the initial region. The method is considered effective only when the image texture is simple or the initial region is considerably accurate. The proposed method, even if the initial region shape is rough and the image texture is complex, can correctly extract the object area by introducing a brightness histogram as a probability density function and an object shape distribution map as a priori probability for the preprocess of the kernel density estimator. It was confirmed that the accuracy and computation speed in general cases are improved in comparison to those of existing method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) object extraction histogtam / kernel density estimator / joint probability density function
Paper # IE2007-265,PRMU2007-249
Date of Issue

Conference Information
Committee IE
Conference Date 2008/3/3(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Image Engineering (IE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Extraction of Object Regions Using Shape Map and Local Features
Sub Title (in English)
Keyword(1) object extraction histogtam
Keyword(2) kernel density estimator
Keyword(3) joint probability density function
1st Author's Name Koji IMAEDA
1st Author's Affiliation Graduate School of Information Science, Nagoya University()
2nd Author's Name Miyako BABA
2nd Author's Affiliation Toyota Central R&D Labs., Inc.
3rd Author's Name Nobuyuki SHIRAKI
3rd Author's Affiliation Toyota Central R&D Labs., Inc.
4th Author's Name Akihiro WATANABE
4th Author's Affiliation Toyota Central R&D Labs., Inc.
5th Author's Name Yasushi HIRANO
5th Author's Affiliation Information Technology Center, Nagoya University
6th Author's Name Shoji KAJITA
6th Author's Affiliation Information Technology Center, Nagoya University
7th Author's Name Kenji MASE
7th Author's Affiliation Information Technology Center, Nagoya University
Date 2008-03-10
Paper # IE2007-265,PRMU2007-249
Volume (vol) vol.107
Number (no) 538
Page pp.pp.-
#Pages 6
Date of Issue