An Effective Deep Neural Network in Edge Computing Enabled Internet of Things for Plant Diseases Monitoring

Yao-Hong Tsai, Tse-Chuan Hsu; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, 2024, pp. 695-699

With the rise of the Internet of Things technology, smart agriculture and the corresponding technology, namely, Taiwan 's agriculture 4.0 has been developed. A large number of large-scale planted agricultural areas have been created in recent years. Unmanned aircraft traveling over the farmland combined with widely distributed smart sensors monitor crop growth, and manage plant disease. Among them, orchids are very suitable as ornamental flowers, and they are flowers with high economic value. Therefore, orchids have become one of the most important agricultural products for export in Taiwan. However, the environment for planting orchids and disease control are very important to prevent the diseases. If orchids become infected, they must immediately make a correct diagnosis of the disease in order to effectively find the way of prevention and treatment measures and minimize the loss. This project intends to study the problem about "An effective deep neural network in edge computing enabled internet of things for plant diseases monitoring", and proposes an effective method for detecting and identifying orchid diseases. It will integrate local and global features in the disease symptoms to the disease attribute and type learning, respectively. Under the automated planting IoT environment, real-time surveillance images are used to identify orchid diseases, and the system can improve identification performance by integrating deep learning neural networks.