Smart Harvesting Decision System for Date Fruit Based on Fruit Detection and Maturity Analysis Using YOLO and K-Means Segmentation

Abstract

The date palm (Phoenixdactylifera) is a large palm with exotic fruits measuring up to 30 metersin height. The date palm produces fruits rich in nutrients provides a multitudeof secondary products, and generates income necessary for the survival of alarge population. Losses attributed to manual harvesting encompass bothquantitative and qualitative aspects, with the latter measured throughattributes such as appearance, taste, texture, and nutritional or economicvalue. These losses, in terms of both quantity and quality, are influenced bypractices across all phases of the harvesting process. On the other hand, therisks of work accidents are high because of the length of the date palms. Toreduce the losses and reduce risks, it is essential to propose a decisionsystem for robotic harvesting to help farmers overcome the constraints duringthe harvest. The assessment of quality and maturity levels in variousagricultural products is heavily reliant on the crucial attribute of color. Inthis study, an intelligent harvesting decision system is proposed to estimatethe level of maturity based on deep learning, K-means clustering, and coloranalysis. The decision system's performance is assessed using the dataset ofdate fruit in the orchard and various metrics. Based on the experimentalresults, the proposed approach has been deemed effective and the systemdemonstrates a high level of accuracy. The system can detect, locate, andanalyze the maturity stage to make a harvest decision.