|
Abstract
|
Article Information:
Classification of Scenes into Indoor/Outdoor
R. Raja, S. Md. Mansoor Roomi and D. Dharmalakshmi
Corresponding Author: R. Raja
Submitted: August 19, 2014
Accepted: September 25, 2014
Published: December 05, 2014 |
Abstract:
|
Effective model for scene classification is essential, to access the desired images from large scale databases. This study presents an efficient scene classification approach by integrating low level features, to reduce the semantic gap between the visual features and richness of human perception. The objective of the study is to categorize an image into indoor or outdoor scene using relevant low level features such as color and texture. The color feature from HSV color model, texture feature through GLCM and entropy computed from UV color space forms the feature vector. To support automatic scene classification, Support Vector Machine (SVM) is implemented on low level features for categorizing a scene into indoor/outdoor. Since the combination of these image features exhibit a distinctive disparity between images containing indoor or outdoor scenes, the proposed method achieves better performance in terms of classification accuracy of about 92.44%. The proposed method has been evaluated on IITM- SCID2 (Scene Classification Image Database) and dataset of 3442 images collected from the web.
Key words: Color model, content based image retrieval, entropy with UV, image annotation, image retrieval, scene classification,
|
|
Abstract
|
PDF
|
HTML |
|
Cite this Reference:
R. Raja, S. Md. Mansoor Roomi and D. Dharmalakshmi, . Classification of Scenes into Indoor/Outdoor. Research Journal of Applied Sciences, Engineering and Technology, (21): 2172-2178.
|
|
|
|
 |
ISSN (Online): 2040-7467
ISSN (Print): 2040-7459 |
 |
| Information |
|
|
|
|
| Sales & Services |
|
|
|
|
