A similar image determination model concerning integrated image features

This model includes Image Text-based Data Analysis (ITDA) module, Image Characteristic-based Data Analysis (ICDA) module and Image Similarity Determination (ISD) module. First, the ITDA module uses information from tags, annotations, and names of the images, and applies text mining technology to ana...

Full description

Saved in:
Bibliographic Details
Published in:Journal of industrial and production engineering 2013-09, Vol.30 (6), p.381-396
Main Authors: Yang, Shih-Ting, Chung, En-Tzu, Tsai, Yao-Chang, Chen, Yu-You
Format: Article
Language:English
Subjects:
Categories
Data analysis
Data mining
Data processing
image features
Image retrieval
image search
keyword extraction
Modules
Production engineering
Search engines
Searching
Similarity
text mining
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This model includes Image Text-based Data Analysis (ITDA) module, Image Characteristic-based Data Analysis (ICDA) module and Image Similarity Determination (ISD) module. First, the ITDA module uses information from tags, annotations, and names of the images, and applies text mining technology to analyze the related keywords, in order to determine the category of the images to serve as the basis for the latter image retrieval process. The ICDA module consists of an analysis of four types of characteristics of the image (e.g., color, shape, space and outline) by incorporating the image clusters. Finally, the ISD module consists of the integrated analytic outcomes of the two modules to find the most significant image with the highest similarities for the searcher's references. In addition to the proposed model, a Similar Image Determination system is also established. The system is also validated on its practicality. In conclusion, the methodology proposed in this paper can quickly provide the searchers with compatible target images, and hence, increase the effectiveness of the image searching process.
ISSN:2168-1015
2168-1023
DOI:10.1080/21681015.2013.844735