| Title |
Deep learning-based multi-spectral identification of grey mould |
| ID_Doc |
65132 |
| Authors |
Giakoumoglou, N; Pechlivani, EM; Sakelliou, A; Klaridopoulos, C; Frangakis, N; Tzovaras, D |
| Title |
Deep learning-based multi-spectral identification of grey mould |
| Year |
2023 |
| Published |
|
| DOI |
10.1016/j.atech.2023.100174 |
| Abstract |
Early detection of economically important plant diseases, such as grey mould caused by Botrytis cinerea, is of major importance for the timely application of disease management strategies and the reduction of impacts on crop production and the environment. In this study, artificial inoculation of leaves of cucumber plants with B. cinerea under controlled environment was performed. Multi-spectral imaging was used to capture the fungal spectrum response at 460, 540, 640, 700, 775 and 875 nm, leveraging both RGB and Near Infrared (NIR) channels. Two annotated image datasets were created from the collected multi-spectral images named Botrytis-detection and Botrytis-classification. Several deep learning-based classification and object detection experiments were conducted on both datasets. Classification results indicated that deep learning models can separate the two classes with accuracy 0.93 (F1-score 0.89), while object detection achieved a mean average precision (mAP(50)) of 0.88, paving the way for future early detection of grey mould caused by B. cinerea. |
| Author Keywords |
Botrytis cinerea; Cucumber; Classification; Object detection; Dataset; Deep learning |
| Index Keywords |
Index Keywords |
| Document Type |
Other |
| Open Access |
Open Access |
| Source |
Emerging Sources Citation Index (ESCI) |
| EID |
WOS:001133868000001 |
| WoS Category |
Agricultural Engineering; Agriculture, Multidisciplinary; Agronomy |
| Research Area |
Agriculture |
| PDF |
https://doi.org/10.1016/j.atech.2023.100174
|