![]() ![]() In particular, detecting disease of plants using CNN algorithms were preferred rather than other deep learning algorithms 22. Recently, these CNN algorithms have been used to develop various tools or programs for the detection or assignment of objectives in various fields 21. However, each architecture has its own unique characteristics and appropriate architectures are required for individual datasets 19, 20. CNN architecture development has focused on improving accuracy or efficiency. These are pre-trained CNN models whose performance has been confirmed. AlexNet 13, VGG19 14, GoogLeNet 15, ResNet 16, and EfficientNet 17 are pre-trained CNN models created by changing the number, composition, arrangement, or calculation method of three types of layers, and ranked high in competition, the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 18. Three types of layers are arranged and connected differently depending on the model architectures and model performance is affected by this architecture 12. Convolutional Neural Network (CNN) is a specialized method to recognize or assign images and consists of fully connected layers, numerous convolution layers, and pooling layers 11. Nodes of individual layers are connected to nodes of adjacent layers. ANN contains a three component processing unit consisting of input, hidden, and output layers 10. Various studies have been carried out to apply deep learning algorithms more precisely to disease detection, such as applying newly developed architectures 5, 6, automatically detecting and classifying lesions in plant images 7, or conducting research on preprocessing methods for incomplete images 8 for practical use.Īrtificial Neural Network (ANN) is an AI technology with an analytic system inspired by the nerve system of the human brain that mimics the way the brain processes information 9. Continued development of improved classification models, such as disease detection, or plant health monitoring, may enable AI-supported decision-making systems for smart agriculture 4. ![]() Therefore, research on combining and applying new technologies to efficiently detect diseases has been conducted, and recently, research on detecting plant diseases in leaves using artificial intelligence (AI) is in progress 3. ![]() Detecting and protecting crops from pathogens is labor-intensive and time-consuming, making it virtually impossible for humans to analyze each plant 2. Although research into the causes and effective treatments for crop diseases is actively underway, monitoring plant health and early detection of pathogens are critical to reduce disease spread and facilitate effective management 1. Our model has the potential to apply to smart farming of Solanaceae crops and will be widely used by adding more various crops as training dataset.Ĭrop disease management is important in agriculture to increase yield and quality by reducing the economic and aesthetic damage caused by plant diseases. The low accuracy of non-model crops was improved by adding these crops to the training dataset implicating expendability of the model. In the validation test, the disease detection model classified crops and disease types with high accuracy (97.09%). The ‘unknown’ is added into categories to generalize the model for wide application. The disease detection model consists of three step classification models, crop classification, disease detection, and disease classification. To develop this system, construction of a stepwise disease detection model using images of diseased-healthy plant pairs and a CNN algorithm consisting of five pre-trained models. Thus, an automated system for disease detecting in crops will play an important role in agriculture by constructing early detection system of disease. However, detection of disease needs specialized knowledge and long-term experiences in plant pathology. Accurately detecting disease occurrences of crops in early stage is essential for quality and yield of crops through the decision of an appropriate treatments. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |