An Edge Computing-Based Solution for Real-Time Leaf Disease Classification using Thermal Imaging
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910686740742144 |
|---|---|
| author | da Silva, Públio Elon Correa Almeida, Jurandy |
| author_facet | da Silva, Públio Elon Correa Almeida, Jurandy |
| contents | Deep learning (DL) technologies can transform agriculture by improving crop health monitoring and management, thus improving food safety. In this paper, we explore the potential of edge computing for real-time classification of leaf diseases using thermal imaging. We present a thermal image dataset for plant disease classification and evaluate deep learning models, including InceptionV3, MobileNetV1, MobileNetV2, and VGG-16, on resource-constrained devices like the Raspberry Pi 4B. Using pruning and quantization-aware training, these models achieve inference times up to 1.48x faster on Edge TPU Max for VGG16, and up to 2.13x faster with precision reduction on Intel NCS2 for MobileNetV1, compared to high-end GPUs like the RTX 3090, while maintaining state-of-the-art accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_03835 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | An Edge Computing-Based Solution for Real-Time Leaf Disease Classification using Thermal Imaging da Silva, Públio Elon Correa Almeida, Jurandy Computer Vision and Pattern Recognition Deep learning (DL) technologies can transform agriculture by improving crop health monitoring and management, thus improving food safety. In this paper, we explore the potential of edge computing for real-time classification of leaf diseases using thermal imaging. We present a thermal image dataset for plant disease classification and evaluate deep learning models, including InceptionV3, MobileNetV1, MobileNetV2, and VGG-16, on resource-constrained devices like the Raspberry Pi 4B. Using pruning and quantization-aware training, these models achieve inference times up to 1.48x faster on Edge TPU Max for VGG16, and up to 2.13x faster with precision reduction on Intel NCS2 for MobileNetV1, compared to high-end GPUs like the RTX 3090, while maintaining state-of-the-art accuracy. |
| title | An Edge Computing-Based Solution for Real-Time Leaf Disease Classification using Thermal Imaging |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2411.03835 |