Meta-classifier based on boosted approach for object class recognition
Object class recognition deals with the classification of individual objects to a certain class. In images of natural scenes, objects appear in a variety of poses and scales, with or without occlusion. Object class recognition typically involves the extraction, processing and analysis of visual feat...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Praise Worthy Prize
2014
|
| _version_ | 1796972516765335552 |
|---|---|
| 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 |
| collection | UPM |
| description | Object class recognition deals with the classification of individual objects to a certain class. In images of natural scenes, objects appear in a variety of poses and scales, with or without occlusion. Object class recognition typically involves the extraction, processing and analysis of visual features such as color, shape, or texture from an object, and then associating a class label to it. In this study, global shape and local features are considered as discriminative features for object class recognition. Both local and shape features are combined in order to obtain better classification performance for each object class. A meta-classifier framework is proposed as a model for object class recognition. Meta-classifier is used to learn a decision classifier that optimally predicts the correctness of classification of base classifier for each object. In this framework, base classifiers based on boosting approach are trained using the local and global shape features, respectively. Then, these classifiers results are combined as input to the meta-classifier. The results from classification experiments showed that meta-classifier based on boosted approach performs better compared to some state-of- the-art approaches in object class recognition. |
| first_indexed | 2024-03-06T08:29:11Z |
| format | Article |
| id | upm.eprints-34498 |
| institution | Universiti Putra Malaysia |
| last_indexed | 2024-03-06T08:29:11Z |
| publishDate | 2014 |
| publisher | Praise Worthy Prize |
| record_format | dspace |
| spelling | upm.eprints-344982015-12-15T07:06:45Z http://psasir.upm.edu.my/id/eprint/34498/ Meta-classifier based on boosted approach for object class recognition Manshor, Noridayu Abdul Rahiman, Amir Rizaan Raja Mahmood, Raja Azlina Object class recognition deals with the classification of individual objects to a certain class. In images of natural scenes, objects appear in a variety of poses and scales, with or without occlusion. Object class recognition typically involves the extraction, processing and analysis of visual features such as color, shape, or texture from an object, and then associating a class label to it. In this study, global shape and local features are considered as discriminative features for object class recognition. Both local and shape features are combined in order to obtain better classification performance for each object class. A meta-classifier framework is proposed as a model for object class recognition. Meta-classifier is used to learn a decision classifier that optimally predicts the correctness of classification of base classifier for each object. In this framework, base classifiers based on boosting approach are trained using the local and global shape features, respectively. Then, these classifiers results are combined as input to the meta-classifier. The results from classification experiments showed that meta-classifier based on boosted approach performs better compared to some state-of- the-art approaches in object class recognition. Praise Worthy Prize 2014 Article PeerReviewed Manshor, Noridayu and Abdul Rahiman, Amir Rizaan and Raja Mahmood, Raja Azlina (2014) Meta-classifier based on boosted approach for object class recognition. International Review on Computers and Software, 9 (9). pp. 1590-1596. ISSN 1828-6003; ESSN: 1828-6011 http://www.praiseworthyprize.org/jsm/index.php?journal=irecos&page=article&op=view&path[]=16183 10.15866/irecos.v9i9.3058 |
| spellingShingle | Manshor, Noridayu Abdul Rahiman, Amir Rizaan Raja Mahmood, Raja Azlina Meta-classifier based on boosted approach for object class recognition |
| title | Meta-classifier based on boosted approach for object class recognition |
| title_full | Meta-classifier based on boosted approach for object class recognition |
| title_fullStr | Meta-classifier based on boosted approach for object class recognition |
| title_full_unstemmed | Meta-classifier based on boosted approach for object class recognition |
| title_short | Meta-classifier based on boosted approach for object class recognition |
| title_sort | meta classifier based on boosted approach for object class recognition |
| work_keys_str_mv | AT manshornoridayu metaclassifierbasedonboostedapproachforobjectclassrecognition AT abdulrahimanamirrizaan metaclassifierbasedonboostedapproachforobjectclassrecognition AT rajamahmoodrajaazlina metaclassifierbasedonboostedapproachforobjectclassrecognition |