Fusion of global shape and local features using meta-classifier framework.
In computer vision, objects in an image can be described using many features such as shape, color, texture and local features. The number of dimensions for each type of feature has differing size. Basically, the underlying belief from a recognition point of view is that, the more features being used...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
Praise Worthy Prize
2013
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/30671/ http://psasir.upm.edu.my/id/eprint/30671/1/Fusion%20of%20global%20shape%20and%20local%20features%20using%20meta.pdf |
| _version_ | 1848846744594415616 |
|---|---|
| author | Manshor, Noridayu Abdul Rahiman, Amir Rizaan Raja Mahmood, Raja Azlina |
| author_facet | Manshor, Noridayu Abdul Rahiman, Amir Rizaan Raja Mahmood, Raja Azlina |
| author_sort | Manshor, Noridayu |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | In computer vision, objects in an image can be described using many features such as shape, color, texture and local features. The number of dimensions for each type of feature has differing size. Basically, the underlying belief from a recognition point of view is that, the more features being used, the better the recognition performance. However, having more features does not necessarily correlate to better performance. The higher dimensional vectors resulting from fusion might contain irrelevant or noisy features that can degrade classifier performance. Repetitive and potentially useless information might be present which further escalates the 'curse of dimensionality' problem. Consequently, unwanted and irrelevant features are removed from the combination of features. Although this technique provides promising recognition performance, it is not efficient when it comes to computational time in model building. This study proposes meta- classifier framework to ensure all relevant features are not ignored, while maintaining minimal computational time. In this framework, individual classifiers are trained using the local and global shape features, respectively. Then, these classifiers results are combined as input to the meta- classifier. Experimental results have shown to be comparable, or superior to existing state-of-the-art works for object class recognition. |
| first_indexed | 2025-11-15T09:07:35Z |
| format | Article |
| id | upm-30671 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English English |
| last_indexed | 2025-11-15T09:07:35Z |
| publishDate | 2013 |
| publisher | Praise Worthy Prize |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-306712015-10-07T01:36:31Z http://psasir.upm.edu.my/id/eprint/30671/ Fusion of global shape and local features using meta-classifier framework. Manshor, Noridayu Abdul Rahiman, Amir Rizaan Raja Mahmood, Raja Azlina In computer vision, objects in an image can be described using many features such as shape, color, texture and local features. The number of dimensions for each type of feature has differing size. Basically, the underlying belief from a recognition point of view is that, the more features being used, the better the recognition performance. However, having more features does not necessarily correlate to better performance. The higher dimensional vectors resulting from fusion might contain irrelevant or noisy features that can degrade classifier performance. Repetitive and potentially useless information might be present which further escalates the 'curse of dimensionality' problem. Consequently, unwanted and irrelevant features are removed from the combination of features. Although this technique provides promising recognition performance, it is not efficient when it comes to computational time in model building. This study proposes meta- classifier framework to ensure all relevant features are not ignored, while maintaining minimal computational time. In this framework, individual classifiers are trained using the local and global shape features, respectively. Then, these classifiers results are combined as input to the meta- classifier. Experimental results have shown to be comparable, or superior to existing state-of-the-art works for object class recognition. Praise Worthy Prize 2013-09 Article PeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/30671/1/Fusion%20of%20global%20shape%20and%20local%20features%20using%20meta.pdf Manshor, Noridayu and Abdul Rahiman, Amir Rizaan and Raja Mahmood, Raja Azlina (2013) Fusion of global shape and local features using meta-classifier framework. International Review on Computers and Software, 8 (9). pp. 2113-2117. ISSN 1828-6003 http://www.praiseworthyprize.org/latest_issues/IRECOS-latest/IRECOS_vol_8_n_9.html English |
| spellingShingle | Manshor, Noridayu Abdul Rahiman, Amir Rizaan Raja Mahmood, Raja Azlina Fusion of global shape and local features using meta-classifier framework. |
| title | Fusion of global shape and local features using meta-classifier framework. |
| title_full | Fusion of global shape and local features using meta-classifier framework. |
| title_fullStr | Fusion of global shape and local features using meta-classifier framework. |
| title_full_unstemmed | Fusion of global shape and local features using meta-classifier framework. |
| title_short | Fusion of global shape and local features using meta-classifier framework. |
| title_sort | fusion of global shape and local features using meta-classifier framework. |
| url | http://psasir.upm.edu.my/id/eprint/30671/ http://psasir.upm.edu.my/id/eprint/30671/ http://psasir.upm.edu.my/id/eprint/30671/1/Fusion%20of%20global%20shape%20and%20local%20features%20using%20meta.pdf |