C052

Ripeness Detection of Oil Palm Fresh Fruit Bunch on Tree

Mohammadmahdi Naghipour, Lew Sook Ling, Tee Connie

AFFILIATION
Faculty of Information Science & Technology, Multimedia University

Description of Invention

The palm oil industry is a huge contributor to Malaysia’s economic growth. Adopting new technologies such as AI to optimise the different processes involved in the palm oil industry will assist the economic growth and will help reduce negative environmental effects. This research aims to develop an interface that can detect ripe FFB from unripe FFB using AI. The YOLOv7 model has been chosen for performing FFB ripeness detection. YOLOv7 can perform the task with an mAP (mean average precision) of 0.94.