An Adequate Image Retrieval Technique Based on Global Level Feature Extraction
Sumaira M.Hayat Khan1, Ayyaz Hussain2, Imad Fakhri Taha Alshaikhli3

1Sumaira M.Hayat Khan, International Islamic University Islamabad, Pakistan.
2Ayyaz Hussain, Asia Pacific University, Malaysia.
3Imad Fakhri Taha Alshaikhli, International Islamic University Malaysia, Selangor, Malaysia. 

Manuscript received on 15 February 2017 | Revised Manuscript received on 22 February 2017 | Manuscript Published on 28 February 2017 | PP: 176-185 | Volume-6 Issue-3, February 2017 | Retrieval Number: C4861026317/17©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Efficient and effective methods are required for the retrieval of relevant data from data stores. The two main approaches for retrieving a required image from a database are known as the local approach and the global approach. This paper presents the technique based on global approach of image feature extraction and comparison. Image features are calculated by taking into account image as a whole. All the three rudimentary image features like; color, texture and shape are utilized in the process of feature vector calculation. Besides these basic image features, Edge Histogram and Fourier Descriptors are also computed to extract edge information and shapes of the objects in the image respectively. Similarity between two images is determined by calculating Euclidean distance between their feature vectors. The experiments in this study were performed on natural images of diverse semantics from a Corel image database, and showed obvious improvement in results compared to several noble systems in the literature.
Keywords: Content Based Image Retrieval, Feature Extraction, Feature Vector, Similarity Measure, Fourier Descriptor, Edge Histogram Descriptor.

Scope of the Article: Information Retrieval