To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Automated Recognition Model for Identifying Harmful and Harmless Insects in Crop Management |
|
Author |
Amal Alshahrani, Rana Alsaedi, Ameera Alfadli, Taif Alahmadi, Deema Alqthami, and Ohoud Alzubaidi |
| Citation |
Vol. 25 No. 5 pp. 69-78 |
|
Abstract |
Agriculture is often affected by the spread of diseases and pests, which can cause significant economic losses. Worldwide, pests can cause yield losses of up to 40%. To minimize these losses, it is crucial to detect and identify pests as early as possible. Prior studies have developed detection models to either detect harmful insects or only harmless insects. However, there has been no model developed to detect both categories together. To address this issue, we aim to develop a model that can detect and classify both harmful and harmless insects in agricultural environments. We will assess the accuracy of three different methods: YOLO (You Only Look Once) versions 8 and 9, and VGG16 (Visual Geometry Group) on a dataset comprising ten classes, five for harmful insects and five for harmless insects, to determine the most effective approach. The results indicate that YOLOv9 achieved the highest accuracy of 0.972, followed closely by YOLOv8 with 0.969, while VGG16 lagged at 0.849. This suggests that YOLOv9 is the most effective tool among the tested models for detecting and classifying both harmful and harmless insects in agricultural settings. |
|
Keywords |
YOLO, VGG-16, Insects, harmful, harmless |
|
URL |
