Implementation YOLOv3 for symptoms of disease in shallots crop

The main agricultural commodity in Indonesia is shallots. This commodity is highly affected by weather conditions. Climate change causes these weather conditions to have an impact on disease. Controlling diseases on shallots quickly and precisely will help farmers in dealing with pests and diseases...

Full description

Saved in:
Bibliographic Details
Main Authors: Sumarudin, A., Suheryadi, Adi, Puspaningrum, Alifiah, Fauza, Aditya Rifqy
Format: Conference Proceeding
Language:English
Subjects:
Agricultural commodities
Algorithms
Commodities
Deep learning
Disease
Disease control
Machine learning
Object recognition
Pests
Signs and symptoms
Weather
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The main agricultural commodity in Indonesia is shallots. This commodity is highly affected by weather conditions. Climate change causes these weather conditions to have an impact on disease. Controlling diseases on shallots quickly and precisely will help farmers in dealing with pests and diseases to avoid crop failure. Deep learning-based pest detection for the detection of disease symptoms from horticultural crops has developed. The detection of these symptoms provides predictions about the diseases present in shallots. This study proposes the detection of shallot disease using YOLOv3 which is an object detection algorithm based on deep learning. This research includes dataset selection, training process, and determination of detection model. symptom detection in this study, divided into 3 classes of disease symptoms. The yolov3 parameters observed were GIoU, objectless, loss classification, precision, recall, mAp, F1 score. Based on the results of the study, it was found that the precision level of yolov3 for the detection of onion disease symptoms was 59.4%. These results are good for the detection of shallot symptoms using deep learning.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0113748