Name
Abstract | ||
|---|---|---|
The agricultural sector has recognized that for crop management to thrive, acquiring relevant information on plants is needed. However, studies have shown that agricultural problems remain difficult in many parts of the world due to the lack of the necessary infrastructures. Using a public dataset of 4, 234 plant images from Aarhus University Signal Processing group in collaboration with University of Southern Denmark, that consist of descriptions under a controlled condition concerning camera radiance and stabilization. This paper uses a convolutional neural network for training and does data augmentation to identify 12 plant species using a variety of image transforms: resize, rotate, flip, scaling and histogram equalization. The trained model achieved an accuracy categorization of 99.74% during validation and 99.69% during testing, with specificities and sensitivities of 99%. In future works, we plan to utilize the model by training it to other types of plants like herbal-medicinal plants and other crops in other countries. Moreover, the proposed method can be integrated into a mobile application with the goal to provide farmers efficient farming practices. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/tencon.2018.8650178 | TENCON IEEE Region 10 Conference Proceedings |
Keywords | Field | DocType |
Classification,deep learning,plant seedling images,convolutional neural network | Categorization,Signal processing,Computer science,Convolutional neural network,Feature extraction,Control engineering,Agriculture,Artificial intelligence,Deep learning,Histogram equalization,Machine learning,Plant species | Conference |
ISSN | Citations | PageRank |
2159-3442 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Catherine R. Alimboyong | 1 | 0 | 0.34 |
| Alexander A. Hernandez | 2 | 3 | 4.49 |
| Ruji P. Medina | 3 | 1 | 5.79 |