Development of a monitoring system for grain loss of paddy rice based on a decision tree algorithm

China has the world’s largest planting area of paddy rice, but large quantities of paddy rice fall to the ground and are lost during harvesting with a combine harvester. Reducing grain loss is an effective way to increase production and revenue. In this study, a monitoring system was developed to monitor the grain loss of the paddy rice and this approach was tested on the test bench for verifying the precision. The development of the monitoring system for grain loss included two stages: the first stage was to collect impact signals using a piezoelectric film, extract the four features of Root Mean Square, Peak number, Frequency and Amplitude (fundamental component), and identify the kernel impact signals using the J48 (C4.5) Decision Tree algorithm. In the second stage, the precision of the monitoring system was tested for the paddy rice at three different moisture contents (10.4%, 19.6%, and 30.4%) and five different grain/impurity ratios (1/0.5, 1/1, 1/1.5, 1/2, and 1/2.5). According to the results, the highest monitoring accuracy was 99.3% (moisture content 30.8% and grain/impurity ratio 1/2.5), the average accuracy of the monitoring tests was 92.6%, and monitoring of grain/impurity ratios between 1/1 and 1/1.5 (>95.4%) had higher accuracy than monitoring the other grain/impurity ratios. Monitoring accuracy decreased as impurities increased. The lowest accuracy for grain loss monitoring was obtained when the grain/impurity ratio was 1/2.5, with monitoring accuracies of 88.2%, 75.7% and 78.8% at moisture contents of 10.4%, 19.6% and 30.4%. Keywords: monitoring system, combine harvester, paddy rice, grain loss, sensor, data mining, decision tree, development DOI: 10.25165/j.ijabe.20211401.5731 Citation: Lian Y, Chen J, Guan Z H, Song J. Development of a monitoring system for grain loss of paddy rice based on a decision tree algorithm. Int J Agric & Biol Eng, 2021; 14(1): 224–229.

[1]  Josse De Baerdemaeker,et al.  Identification of the cleaning process on combine harvesters. Part I: A fuzzy model for prediction of the material other than grain (MOG) content in the grain bin , 2008 .

[2]  Josse De Baerdemaeker,et al.  Identification of the cleaning process on combine harvesters, Part II: A fuzzy model for prediction of the sieve losses , 2010 .

[3]  J. De Baerdemaeker,et al.  An on-the-go monitoring algorithm for separation processes in combine harvesters , 2004 .

[4]  Josse De Baerdemaeker,et al.  Fuzzy control of the cleaning process on a combine harvester , 2010 .

[5]  Yaoming Li,et al.  Optimum design of grain impact sensor utilising polyvinylidene fluoride films and a floating raft damping structure , 2012 .

[6]  Jin Chen,et al.  Real-time grain breakage sensing for rice combine harvesters using machine vision technology , 2020 .

[7]  Soyoung Park,et al.  Seismic Vulnerability Assessment and Mapping of Gyeongju, South Korea Using Frequency Ratio, Decision Tree, and Random Forest , 2020, Sustainability.

[8]  A. Larsolle,et al.  A Spatially Explicit Decision Support System for Assessment of Tree Stump Harvest Using Biodiversity and Economic Criteria , 2020 .

[9]  Saro Lee,et al.  Novel Credal Decision Tree-Based Ensemble Approaches for Predicting the Landslide Susceptibility , 2020, Remote. Sens..

[10]  Yaoming Li,et al.  Real-time grain impurity sensing for rice combine harvesters using image processing and decision-tree algorithm , 2020, Comput. Electron. Agric..

[11]  Zhong Tang,et al.  Optimum design of an array structure for the grain loss sensor to upgrade its resolution for harvesting rice in a combine harvester , 2017 .

[12]  Chen Yi,et al.  Structure optimization and performance experiment of grain loss monitoring sensor in combine harvester in using modal analysis , 2013 .

[13]  Zhan Zhao,et al.  Structure Optimization of a Grain Impact Piezoelectric Sensor and Its Application for Monitoring Separation Losses on Tangential-Axial Combine Harvesters , 2015, Sensors.

[14]  Li Yang,et al.  Optimum design of grain sieve losses monitoring sensor utilizing partial constrained viscoelastic layer damping (PCLD) treatment , 2015 .

[15]  Jin Chen,et al.  A fuzzy control strategy for the forward speed of a combine harvester based on KDD , 2017 .

[16]  Yaoming Li,et al.  Original papers: Grain separation loss monitoring system in combine harvester , 2011 .

[17]  S. Narendranath,et al.  Fault Diagnosis of Face Milling Tool using Decision Tree and Sound Signal , 2018 .

[18]  Xiaodong Liu,et al.  A Pearson's correlation coefficient based decision tree and its parallel implementation , 2018, Inf. Sci..