Plant disease detection based on lightweight CNN model
- Resource Type
- Conference
- Authors
- Liu, Yang; Gao, Guoqin; Zhang, Zhenhui
- Source
- 2021 4th International Conference on Information and Computer Technologies (ICICT) ICICT Information and Computer Technologies (ICICT), 2021 4th International Conference on. :64-68 Mar, 2021
- Subject
- Computing and Processing
Training
Embedded systems
Computational modeling
Sociology
Training data
Production
Computational complexity
plant disease detection
deep learning
transfer learning
light-weight CNN
- Language
The human population keeps increasing over the last decades. It requires a significant increase in agricultural production. However, the agricultural production is greatly affected by various plant diseases. Timely and accurately identifying the types of leaf diseases is very important for plant diseases control. Convolutional neural network (CNN) is one of the most popular ways for image identification. It can automatically learn appropriate features from training data. In this paper, we propose a light-weight CNN model based on SqueezeNet. The proposed model is trained and tested using the open source PlantVillage dataset. Testing results show that the proposed model can achieve an accuracy of 98.46% while the memory requirements for this model is only 0.62 MB. This demonstrates the technical feasibility of light-weight CNNs in classifying plant diseases using embedded system.