Comparison of palm oil Fresh Fruit Bunches (FFB) ripeness classification technique using deep learning method

• Main Authors: Khamis, Nurulaqilla, Selamat, Hazlina, Ghazalli, Shuwaibatul Aslamiah, Md. Saleh, Nurul Izrin, Yusoff, Nooraini
• Format: Conference or Workshop Item
• Published: 2022
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/98624/; http://dx.doi.org/10.23919/ASCC56756.2022.9828345

The ripeness of palm oil fruit is currently determined through manual visual inspection by palm oil estate workers that could result inconsistent and inaccurate fruit grading. Moreover, the manual inspection is time-consuming and exhausting duty for humans to complete the daily repetitive task. To overcome this issue, this paper proposes an automatic fruit grading classification by utilizing computer vision technologies. A comparison using image classification (ResNet50) and object detection (YOLOv3) technique is analysed in this work. It is clearly demonstrated that object detection model is remarkable in improving ripeness category based on the finer level of feature that has been extracted during the convolutional process.