Potato is one of the world’s four major food crops as an important resource cultivated in about 150 countries. As precision agriculture has recently attracted increasing attention for its role in improving productivity, interest in yield monitoring is also increasing. Yield monitoring is a precision agriculture technology, and it can help farmhouse business management in the future by contributing to variable fertilization and supply and demand control. The present study was carried out to develop and evaluate a system that uses machine vision and deep learning technologies to estimate potato mass to monitor potato yield. The system performs object classification using the YOLOv5 algorithm to sort out potatoes among various foreign substances, object tracking using the DeepSORT algorithm to track the sorted potatoes, and volume calculation using the lengths of the major axis and minor axis of the tracked potatoes. The results of analyzing the function of the developed yield monitoring system showed an object detection rate of 95.2% and a weight measurement error of 9%, indicating that the computation load must be reduced through algorithm optimization to improve the accuracy and that error correction needs to be performed based on the potato position within the view angle.
This study attempted to develop and validate a data-driven simulation model that integrates field-measured data to assess the power requirement and fuel consumption characteristics of a self-propelled collector. The collector is a hydrostatic transmission-based, crawler-type platform designed for garlic and onion harvesting, equipped with multiple hydraulic subsystems for collection and sorting. During field experiments, the power requirements of each subsystem and fuel flow rate were recorded, and Willans line method was applied to estimate engine power and subsystem power transmission efficiencies. Because many small agricultural machines do not support electronically instrumented engines (e.g., CAN-bus/ECU-based measurements), the proposed model was formulated as a data-driven, low-order representation derived from on-site measurements rather than a full physics-based model. Using the identified parameters, the simulation framework predicts engine power and fuel efficiency under various operating conditions. The simulation results exhibited high agreement with field data, achieving R2 and mean absolute percentage error values of 0.935–0.981 and 1.79–4.18%, respectively, confirming reliable reproduction of real field performance. A comprehensive analysis of the simulation results revealed that both engine speed and travel speed significantly influence power distribution and fuel rate, while also indicating that hydraulic working power is the dominant contributor to total power demand at higher engine speeds. These findings provide practical guidance for improving the fuel efficiency of compact self-propelled collectors.
30 December 2021 in Journal of Agriculture & Life Science