Colour texture image classification using colour completed local binary pattern (CCLBP)
Local Binary Pattern (LBP) descriptor is being used successfully for the classification of textures. Also, it is been used for other tasks such as facial expression, face recognition and texture segmentation. On the other hand, these descriptors are barely used for image categorization due to their...
| Main Author: | |
|---|---|
| Format: | Undergraduates Project Papers |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/26531/ http://umpir.ump.edu.my/id/eprint/26531/ http://umpir.ump.edu.my/id/eprint/26531/1/Colour%20texture%20image%20classification%20using%20colour%20completed%20local%20binary%20pattern.pdf |
| Summary: | Local Binary Pattern (LBP) descriptor is being used successfully for the classification of textures. Also, it is been used for other tasks such as facial expression, face recognition and texture segmentation. On the other hand, these descriptors are barely used for image categorization due to their calculations which are depend on the gray image and they are invariant to monotonic light variations on the gray level. Despite the key role in distinctive the objects of these descriptors, they ignore color information. In this project, Completed Local Binary Pattern (CLBP) will be enhanced and two colour CLBP descriptors are proposed which RGB_CCLBP and HSV_CCLBP. Moreover, the datasets that have been used in this project are KTH-TIPS, KTH-TIPS 2A and Outex_TC_00013 datasets. The proposed method shows promising results despite the limitations of it. |
|---|