Presentation 2004/1/26
A Segmentation Method for Real Images using Quadratic Curved Line Units
Toru MIYAKOSHI, Masashi KAMEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We have proposed a new image coding method based on the geometrical characteristics of the image. The proposed method divides an input image into some regions on the 2-dimensional spatial domain, and each region is approximated by the quadratic curved line units. In this paper, we have presented the method to determine the length of each region in the viewpoint of the minimum square error, and we have shown the coding efficiency by our proposed method when the input image is given. In our experimental results, the reconstructed image by our proposed method doesn't have remarkable distortions compared with the same quality JPEG image.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Coding / Geometrical Characteristics / Image Segmentation / Quadratic Spline Curve / Method of Least Squares
Paper # ITS2003-51,IE2003-186
Date of Issue

Conference Information
Committee ITS
Conference Date 2004/1/26(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 Intelligent Transport Systems Technology (ITS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Segmentation Method for Real Images using Quadratic Curved Line Units
Sub Title (in English)
Keyword(1) Image Coding
Keyword(2) Geometrical Characteristics
Keyword(3) Image Segmentation
Keyword(4) Quadratic Spline Curve
Keyword(5) Method of Least Squares
1st Author's Name Toru MIYAKOSHI
1st Author's Affiliation Graduate School of Software and Information Science, Iwate Prefectural University()
2nd Author's Name Masashi KAMEDA
2nd Author's Affiliation Graduate School of Software and Information Science, Iwate Prefectural University
Date 2004/1/26
Paper # ITS2003-51,IE2003-186
Volume (vol) vol.103
Number (no) 639
Page pp.pp.-
#Pages 6
Date of Issue