Average Analysis Method in Selecting Haralick's Texture Features on Color Co-occurrence Matrix for Texture Based Image Retrieval

Mamat, Abd. Rasid and Awang, Mohd Khalid and Abdul Rawi, Norkhairani and Awang, Mohd Isa and Abdul Kadir, Mohd Fadzil (2016) Average Analysis Method in Selecting Haralick's Texture Features on Color Co-occurrence Matrix for Texture Based Image Retrieval. International Journal of Multimedia and Ubiquitous Engineering, 11 (2). 00-00. ISSN 1975-0080 (In Press)

[img] Text
10 IJMUE Rasid.pdf
Restricted to Registered users only

Download (736Kb) | Request a copy
Official URL: http://www.sersc.org


Many textures based image retrieval researchers use global texture features for representing and retrieval of images from an image database. Generally, such researches suffer from the misrepresentation of local information leading to the inefficient image retrieval performance. This paper focuses on extracting local Haralick’s texture feature based on a predetermined region using the color co-occurrence matrix method, the selection of the ‘contributed’ Haralik’s texture features and evaluated performance of the combination of ‘contributed’ features. In this research we have utilized two methods of ‘contributed’ feature selection. Firstly, the proposed method, namely the Average Analysis (AA) of Precision and Recall and well known method, Principal Component Analysis (PCA). In order to compare the performance, a series of experiments were carried out on the findings of AA and PCA methods. Experiments were made based on 1000 selected images from the Coral image database which were divided into ten categories. From the experimental findings, it is interesting to note that for the combination features, the resulting from the proposed method (AA) showed better retrieval performance compared to the PCA method for almost all categories. This finding has an important implication in deciding the correct combination of ‘contributed’ features for certain image properties. It has shown that the proposed method is able to produce less computational processing time due to less processing involved. The result is also compared to the previous researchers and shown an increase of average precision from 8.5% to 26%.

Item Type: Article
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / Institute: Faculty of Informatics & Computing
Depositing User: Norkhairani Abdul Rawi
Date Deposited: 02 May 2016 02:47
Last Modified: 02 May 2016 02:47
URI: http://erep.unisza.edu.my/id/eprint/4060

Actions (login required)

View Item View Item