|
Sugar Cane Red Stripe Disease Detection using YOLO CNN of Deep Learning Technique |
|---|---|
| รหัสดีโอไอ | |
| Creator | Inchaya Kumpala |
| Title | Sugar Cane Red Stripe Disease Detection using YOLO CNN of Deep Learning Technique |
| Contributor | Nontawat Wichapha, Piyawat Prasomsab |
| Publisher | Faculty of Engineering Mahasasakham University |
| Publication Year | 2565 |
| Journal Title | Engineering Access |
| Journal Vol. | 8 |
| Journal No. | 2 |
| Page no. | 192-197 |
| Keyword | YOLO, image processing, deep learning, Convolutional Neural Network, CNN, sugar cane, leaf-disease, image classification |
| URL Website | https://ph02.tci-thaijo.org/index.php/mijet/index |
| Website title | THAIJO Engineering Access |
| ISSN | 2730-4175 |
| Abstract | The objective of this research is to apply the deep learning technology based on the Convolutional Neural Network (CNN) algorithm YOLO, creating a simulation for image recognition. The technology was used to recognise the sugar cane disease with specified images. The Sugar cane-Leaf Disease Diagnosis System was designed and developed to enable the user to recognise sugar cane disease automatically. Sugar cane-Leaf Disease Diagnosis System consisted of two parts: the first part was the disease detection and diagnosis. This was where the Convolutional Neural Network learning-teaching import 4,000 images divided into 2,000 images of sugar cane leaves with disease and 2,000 images of sugar cane without disease for the comparison. The other part was the system for displaying response or disease diagnosis system interface. This part contained the Convolutional Neural Network used to categorize and analyzing the leaf condition that would be diseased and non-diseased. The tool used for sugar cane leaf recognition and analysis in this research was the Deep Learning technique based on a Convolutional Neural Network consisting of image classification, image analysis, and image processing. This tool was used to test 3 sample groups, which were selected from 9 promotional staff from Mitrphol sugar factory, Thailand, 3 operative agricultural academic experts from Khon Kaen Field Crop Research Center, Thailand, 2 system developers, and 30 local agriculturists. The average accuracy score of processing of the first and the second group was 95.90 % and 91.30% with the highest accuracy of 98.45% and 97.26%, respectively, while the average estimated time duration was 1.46 and 1.53 seconds, respectively. |
