Classification of Oil Palm Trees Using Quantum Convolutional Neural Network (QCNN)

Abstract

The oil palm trees (Elaeis guineensis) is an important commodity in the plantation industry, and the classification
of its varieties is crucial for enhancing productivity and harvesting efficiency. This study aims to apply Quantum
Convolutional Neural Network (QCNN) as a method for classifying oil palm trees. QCNN integrates quantum computing
principles into the architecture of convolutional neural networks, allowing for more efficient and accurate data processing.
The data used in this research includes oil palm plantations located in the coastal areas of Bengkalis Island. Data acquisition
was performed using a DJI Phantom 4 Pro drone, capturing vertical images from above. The classification process utilized
key features extracted from the images using the QCNN algorithm. The results of the experiments show that QCNN
achieved a training accuracy of 94.7% and a testing accuracy of 92.5%. Thus, this research makes a significant contribution
to the development of oil palm tree classification technology and opens new opportunities for the application of quantum
algorithms in agriculture. These findings are expected to assist farmers and researchers in identifying and managing oil
palm tree varieties more effectively, thereby supporting the sustainability and productivity of the oil palm industry as a
whole.