Please use this identifier to cite or link to this item:
https://dspace.univ-adrar.edu.dz/jspui/handle/123456789/1506
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | ABAIDI, Abedelkader | |
dc.contributor.author | Aymen, Mohamed | |
dc.contributor.author | Kohili, Mohammed / Supervisor | |
dc.date.accessioned | 2019-06-10T18:37:36Z | |
dc.date.available | 2019-06-10T18:37:36Z | |
dc.date.issued | 2018 | |
dc.identifier.uri | http://www.univ-adrar.dz/:8080/xmlui/handle/123456789/1506 | |
dc.description | Computer Science | en_US |
dc.description.abstract | We present in this thesis an experimental study, which aims to do an optimal selection of SVM model for the recognition of leaves. Include the generic Fourier descriptor for features extraction and type of classification (one-against-all) the choice of these parameters has a great influence on the final performance of the classification and the computation time. This work required the realization of a complete system of recognition offline of leaves, the generation of database of 32 classes each class has more than 50 images, which is approximately 1907 images. Preliminary results are very encouraging and promising compared to the general literature survey. | en_US |
dc.language.iso | en | en_US |
dc.publisher | Ahmed Draia University - Adrar | en_US |
dc.subject | SVM multiclass | en_US |
dc.subject | Generic Fourier Discriptor GFD | en_US |
dc.subject | Recognition | en_US |
dc.subject | leaves | en_US |
dc.subject | Support Vector machine SVM | en_US |
dc.subject | Intelligent Systems | en_US |
dc.title | Recognition of Leaf Images Based on generic fourier descriptor Using SVM | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | Mémoires de Master |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Recognition of Leaf Images Based on generic fourier descriptor Using SVM.pdf | 6.48 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.