Spatially Localized Circular and Overlapped Feature Extraction for Gray Scale Images using Gabor Jets

Urolagin, Siddhaling and Prema, KV and Jayakrishna, R (2013) Spatially Localized Circular and Overlapped Feature Extraction for Gray Scale Images using Gabor Jets. International Journal of Computer Applications, 61 (7). pp. 40-44. ISSN 0975 – 8887

[img] PDF
pxc3884583.pdf - Published Version
Restricted to Registered users only

Download (292kB) | Request a copy
Official URL: http://www.ijcaonline.org/

Abstract

Extracting the effective features from the image for object recognition is one of the most important problems in computer vision. One of the usual methods of feature extraction is through binarization process. Binarizing the gray scale image has many disadvantages: vital brightness information loss, difficult to set proper threshold. In this paper feature-extracting method directly from the gray scale images using Gabor jets is discussed. Rather than naïve use of Gabor jets as features, a spatially localized circular and overlapped method for computing the features of Gabor jets to form feature vectors has been proposed. To demonstrate the effectiveness and efficacy of feature extraction method, classification of objects under presence of significant amount distortion using simple classifier such as K-nearest neighbor is done. A good rate of recognition is observed in the experimental results.

Item Type: Article
Uncontrolled Keywords: Vision, Feature Extraction, Gray Scale Images, Gabor Jets
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Depositing User: MIT Library
Date Deposited: 20 Mar 2013 06:01
Last Modified: 20 Mar 2013 06:01
URI: http://eprints.manipal.edu/id/eprint/79185

Actions (login required)

View Item View Item