TRAINING OF YOLOV5 FOR APPLE DETECTION IN  ORCHARD

Aleksejs Kardjalis; Inga  Savicka; Kaspars Rubuļņiks; Sergejs Kodors

doi:10.17770/het2024.28.8259

Authors

Aleksejs Kardjalis Rezekne Academy of Technologies, Rezekne (LV)
Inga Savicka Rezekne Academy of Technologies, Rezekne (LV)
Kaspars Rubuļņiks Rezekne Academy of Technologies, Rezekne (LV)
Dr.sc.ing. Sergejs Kodors Rezekne Academy of Technologies, Rezekne (LV)

DOI:

https://doi.org/10.17770/het2024.28.8259

Keywords:

artificial intelligence, fruits, object detection, precision horticulture, YOLOv5

Abstract

Fruit cultivation is a significant part of the country economics and agriculture. In this paper, aiming to improve apple cultivation, we trained a neural network for apple detection using the YOLOv5 architecture, utilizing the dataset from the lzp-2021/1-0134 project. The dataset consisted of a set of apple tree photographs with apple fruits. Dataset contained 40 images with size 640x640 px. YOLOv5m model was trained five times. The best result model achieved mAP@0.5 equal to 0.9 and mAP@0.5:0.95 equal to 0.63. The artificial intelligence opens new possibilities for horticulture saving resources, which can be redirected on other tasks significantly increasing the efficiency of commercial orchards.