Flower classification using deep convolutional neural networks
Flower classification is a challenging task due to the wide range of flower species which have similar shape, appearance or surrounding objects such as leaves and grass. In this paper, we propose a novel two-step deep learning classifier to distinguish flowers of a wide range of species. Firstly, th...
| Asıl Yazarlar: | , , , |
|---|---|
| Materyal Türü: | Journal article |
| Baskı/Yayın Bilgisi: |
Institution of Engineering and Technology
2018
|
| _version_ | 1826292463983132672 |
|---|---|
| author | Hiary, H Saadeh, H Saadeh, M Yaqub, M |
| author_facet | Hiary, H Saadeh, H Saadeh, M Yaqub, M |
| author_sort | Hiary, H |
| collection | OXFORD |
| description | Flower classification is a challenging task due to the wide range of flower species which have similar shape, appearance or surrounding objects such as leaves and grass. In this paper, we propose a novel two-step deep learning classifier to distinguish flowers of a wide range of species. Firstly, the flower region is automatically segmented to allow localisation of the minimum bounding box around it. The proposed flower segmentation approach is modelled as a binary classifier in a fully convolutional network framework. Secondly, we build a robust convolutional neural network classifier to distinguish the different flower types. We propose novel steps during the training stage to ensure robust, accurate and real-time classification. We evaluate our method on three well known flower datasets. Our classification results exceed 97% on all datasets which is better than the state-of-the-art in this domain. |
| first_indexed | 2024-03-07T03:15:06Z |
| format | Journal article |
| id | oxford-uuid:b586850f-61b8-4183-81de-937ef1750c79 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T03:15:06Z |
| publishDate | 2018 |
| publisher | Institution of Engineering and Technology |
| record_format | dspace |
| spelling | oxford-uuid:b586850f-61b8-4183-81de-937ef1750c792022-03-27T04:34:02ZFlower classification using deep convolutional neural networksJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:b586850f-61b8-4183-81de-937ef1750c79Symplectic Elements at OxfordInstitution of Engineering and Technology2018Hiary, HSaadeh, HSaadeh, MYaqub, MFlower classification is a challenging task due to the wide range of flower species which have similar shape, appearance or surrounding objects such as leaves and grass. In this paper, we propose a novel two-step deep learning classifier to distinguish flowers of a wide range of species. Firstly, the flower region is automatically segmented to allow localisation of the minimum bounding box around it. The proposed flower segmentation approach is modelled as a binary classifier in a fully convolutional network framework. Secondly, we build a robust convolutional neural network classifier to distinguish the different flower types. We propose novel steps during the training stage to ensure robust, accurate and real-time classification. We evaluate our method on three well known flower datasets. Our classification results exceed 97% on all datasets which is better than the state-of-the-art in this domain. |
| spellingShingle | Hiary, H Saadeh, H Saadeh, M Yaqub, M Flower classification using deep convolutional neural networks |
| title | Flower classification using deep convolutional neural networks |
| title_full | Flower classification using deep convolutional neural networks |
| title_fullStr | Flower classification using deep convolutional neural networks |
| title_full_unstemmed | Flower classification using deep convolutional neural networks |
| title_short | Flower classification using deep convolutional neural networks |
| title_sort | flower classification using deep convolutional neural networks |
| work_keys_str_mv | AT hiaryh flowerclassificationusingdeepconvolutionalneuralnetworks AT saadehh flowerclassificationusingdeepconvolutionalneuralnetworks AT saadehm flowerclassificationusingdeepconvolutionalneuralnetworks AT yaqubm flowerclassificationusingdeepconvolutionalneuralnetworks |