Experimental Investigation on Combining Different Haralick’s Texture Features for Efficient Color Image Retrieval

Mamat, Abd. Rasid and Hitam, Mohammad Suzuri and Mat Saman, Mohd Yazid and Hamzah, Mohd Pouzi (2011) Experimental Investigation on Combining Different Haralick’s Texture Features for Efficient Color Image Retrieval. In: 2nd WORLD CONFERENCE on INFORMATION TECHNOLOGY (WCIT-2011), 23-26 , Antalya, Turkey.

[img] Text
Restricted to Registered users only

Download (975Kb) | Request a copy
Official URL: http://www.allconferences.com/conferences/2011/201...


Many texture based image retrieval researches use global texture features for representing and retrieval of images from image database. Generally such researches suffer from misrepresentation of local information leading to inefficient image retrieval performance. This paper focuses on extracting local Haralick’s texture feature based on predetermined region using color co-occurrence matrix method. Extensive experimental investigations were conducted to determine the best combination of eleven Haralick’s texture features that will provide the most efficient image retrieval performance based on precision and recall criterion. Evaluations of the retrieval performance were made based on 1000 selected images from Coral image database. These test images were divided into ten categories. From the experimental findings, it is interesting to note that for certain image categories, only four combination of the eleven Haralick’s texture features namely homogenity, sum of squares and sum average and variance provides the best image retrieval performance. This finding has important implication on the use of correct combination of contributed texture features for certain image properties as well as leading to less computational processing time due to less processing involved. However, more research on the effect of different combination of Haralick’s texture feature for efficient image retrieval, need to be undertaken before the association between these texture features is more clearly understood.

Item Type: Conference or Workshop Item (Paper)
Keywords: Color Co-occurrence matrix, Haralick’s texture features,Texture based image retrieval
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA76 Computer software
Faculty / Institute: Faculty of Informatics & Computing
Depositing User: abd rasid mamat
Date Deposited: 15 Nov 2015 07:44
Last Modified: 15 Nov 2015 07:44
URI: http://erep.unisza.edu.my/id/eprint/4054

Actions (login required)

View Item View Item