A novel multi-head CNN design to identify plant diseases using the fusion of RGB images |
| |
| Institution: | 1. Civil Engineering Department, University of Sistan and Baluchestan, Zahedan, Iran;2. Water Security & Sustainable Development Hub, School of Engineering, Newcastle University, Newcastle Upon Tyne, UK;3. School of Engineering, Newcastle University, Newcastle upon Tyne, UK;4. Chair of Engineering Hydrology and Water Management, Technical University of Darmstadt, Darmstadt, Germany;1. Key Laboratory of Mariculture (Ministry of Education), Fisheries College, Ocean University of China, Qingdao 266003, China;2. Southern Marine Science and Engineering Guangdong Laboratory, Guangzhou 511458, China;1. Plant Ecology Group, Faculty of Mathematics & Natural Sciences, University of Tübingen, Tübingen 72076, Germany;2. Physical Geography and Geoinformatics, Faculty of Mathematics & Natural Sciences, University of Tübingen, Tübingen 72070, Germany;3. Earth Observation and Climate Processes, Faculty VI Spatial and Environmental Sciences, Trier University, Trier 54286, Germany;4. Land and Spatial Sciences, Faculty of Engineering & Built Environment, Namibia University of Science and Technology, Windhoek 9000, Namibia;1. School of Computer Science, China West Normal University, Nanchong 637009, Sichuan, China;2. College of Computer Science, Sichuan University, Chengdu 610041, Sichuan, China |
| |
| Abstract: | Plant diseases and insect pests cause a significant threat to agricultural production. Early detection and diagnosis of these diseases are critical and can reduce economic losses. The recent development of deep learning (DL) benefits various fields, such as image processing, remote sensing, medical diagnosis, and agriculture. This work proposed a novel approach based on DL for plant disease detection by fusing RGB and segmented images. A multi-headed DenseNet-based architecture was developed, considering two images as input. We evaluated the model on a public dataset, PlantVillage, consisting of 54183 images with 38 classes. The fivefold cross-validation technique achieved an average accuracy, recall, precision, and f1-score of 98.17%, 98.17%, 98.16%, and 98.12%, respectively. The proposed approach can distinguish various plant diseases with different characteristics by image fusion. The high success rate with low standard deviation proves the robustness of the model, and the model can be integrated into plant disease detection and early warning system. |
| |
| Keywords: | Plant disease detection Fusion CNN Deep learning DenseNet |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
