Title |
Deep Learning for Image-Based Cassava Disease Detection
|
---|---|
Published in |
Frontiers in Plant Science, October 2017
|
DOI | 10.3389/fpls.2017.01852 |
Pubmed ID | |
Authors |
Amanda Ramcharan, Kelsee Baranowski, Peter McCloskey, Babuali Ahmed, James Legg, David P. Hughes |
Abstract |
Cassava is the third largest source of carbohydrates for human food in the world but is vulnerable to virus diseases, which threaten to destabilize food security in sub-Saharan Africa. Novel methods of cassava disease detection are needed to support improved control which will prevent this crisis. Image recognition offers both a cost effective and scalable technology for disease detection. New deep learning models offer an avenue for this technology to be easily deployed on mobile devices. Using a dataset of cassava disease images taken in the field in Tanzania, we applied transfer learning to train a deep convolutional neural network to identify three diseases and two types of pest damage (or lack thereof). The best trained model accuracies were 98% for brown leaf spot (BLS), 96% for red mite damage (RMD), 95% for green mite damage (GMD), 98% for cassava brown streak disease (CBSD), and 96% for cassava mosaic disease (CMD). The best model achieved an overall accuracy of 93% for data not used in the training process. Our results show that the transfer learning approach for image recognition of field images offers a fast, affordable, and easily deployable strategy for digital plant disease detection. |
X Demographics
Geographical breakdown
Country | Count | As % |
---|---|---|
France | 6 | 14% |
Switzerland | 5 | 12% |
United States | 5 | 12% |
Colombia | 2 | 5% |
Germany | 2 | 5% |
Tanzania, United Republic of | 1 | 2% |
Taiwan | 1 | 2% |
Japan | 1 | 2% |
Netherlands | 1 | 2% |
Other | 5 | 12% |
Unknown | 13 | 31% |
Demographic breakdown
Type | Count | As % |
---|---|---|
Members of the public | 34 | 81% |
Scientists | 8 | 19% |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
Unknown | 648 | 100% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Student > Master | 90 | 14% |
Student > Ph. D. Student | 67 | 10% |
Student > Bachelor | 57 | 9% |
Researcher | 56 | 9% |
Lecturer | 30 | 5% |
Other | 91 | 14% |
Unknown | 257 | 40% |
Readers by discipline | Count | As % |
---|---|---|
Computer Science | 146 | 23% |
Engineering | 96 | 15% |
Agricultural and Biological Sciences | 54 | 8% |
Unspecified | 17 | 3% |
Biochemistry, Genetics and Molecular Biology | 11 | 2% |
Other | 52 | 8% |
Unknown | 272 | 42% |