Representation of clipart image using shape and color with spatial relationship

Authors:
Jeong-Hyun Cho;Chang-Gyu Choi;Yongseok Chang;Sung-Ho Kim
Affiliations:
Division of Computer Technology, Yeungnam College of Science & Technology, Daegu, Korea;Dept. of Computer Engineering, Kyungpook National University, Daegu, Korea;Dept. of Computer Engineering, Kyungpook National University, Daegu, Korea;Dept. of Computer Engineering, Kyungpook National University, Daegu, Korea
Venue:
PCM'04 Proceedings of the 5th Pacific Rim conference on Advances in Multimedia Information Processing - Volume Part I
Year:
2004

Citing 6
Cited 0

Computational geometry: an introduction

Computational geometry: an introduction
Shape measures for content based image retrieval: a comparison

Information Processing and Management: an International Journal
Digital Image Processing

Digital Image Processing
Association and Content-Based Retrieval

IEEE Transactions on Knowledge and Data Engineering
An Effective Content-Based Visual Image Retrieval System

COMPSAC '02 Proceedings of the 26th International Computer Software and Applications Conference on Prolonging Software Life: Development and Redevelopment
A Corner-Finding Algorithm for Chain-Coded Curves

IEEE Transactions on Computers

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a method of extracting shape and color information from clipart images, then measuring the similarity between different clipart images using the extracted shape and color information. To represent the shape of clipart images, the proposed method expresses the convex and concave aspects of an outline using the ratio of a rectangle. The shape outline is then expressed using a rectangle representation and converted into a chain code. Meanwhile, the color is represented by color values quantized into HSV color coordinate, the location information between the color regions, and the size of the color regions. These color information can more appropriate for representing clipart images, which have a smaller number of colors than nature images. Experimental results showed that the proposed method is superior in expressing shape and color information than previous methods.