Comparative analysis of image search algorithm using average RGB, local color histogram, global color histogram and color moment HSV

Image retrieval forms a major problem when a large database is considered. Content Base Image Retrieval (CBIR) makes use of the available visual features of the image and helps in retrieving similar image as that of the query image. In the CBIR method, each image stored in the database has its features extracted and compared to the features of the query image. Thus, it involves two processes, feature extraction and feature matching. In this thesis, four techniques have been used, which are the Average of Red, Green and Blue Color Channels (Average RGB), Local Color Histogram (LCH), Global Color Histogram (GCH) and Color Moment of Hue, Saturation and Brightness Value (HSV) to retrieve relevant images based on colour. These techniques are applied on the collection of three images chosen randomly from each class of Wang images database. The performance of each technique has been individually evaluated, in terms of Execution Time, Precision, Recall, Accuracy, Redundancy Factor and Fall Rate. The results were then analysed and compared. The comparison was shown in bar graphs that the Average RGB technique has the best performance, where it obtained high accuracy. As a conclusion to the report, this comparative study contributes to the image searching field, by measuring the performance for several CBIR techniques using more commonly used parameters.

Name dc:type.qualificationname

mphil

Level dc:type.qualificationlevel

masters

Grantor dc:publisher.institution

Universiti Tun Hussein Onn Malaysia

Year dc:date.issued

2015

Author dc:creator

• H.M Alzoubi, Amera

• TA1501-1820 Applied optics. Photonics

Language dc:language

en