数字农科院2.0

Automatic grain unloading method for track-driven rice combine harvesters based on stereo vision

文献类型： 外文期刊

作者： Cui Z.;Hu J.;Yu Y.;Cao G.;Zhang H.;Chai X.;Chen H.;Xu L.

作者机构：

关键词： Automatic grain unloading;Instance segmentation;Stereo vision;Track-driven rice combine harvester

期刊名称： Computers and Electronics in Agriculture

ISSN： 0168-1699

年卷期： 2024 年 220 卷

页码：

收录情况： SCIE（2024版） ; ; EI（2024版）

摘要： Traditional track-driven rice combine harvesters, during the grain unloading process, often depend on the operator's frequent adjustment of the grain unloader's position while closely monitoring the accumulation of grain within the truck. Because of the structural characteristics of the harvester and the narrowness of the rural terrain, visibility is often obstructed, thereby increasing the difficulty of operation. It is estimated that the time consumed for unloading grain constitutes almost half of the total harvesting time, significantly reducing the efficiency of the harvester. To address these challenges, this study initially proposes an automated grain unloading system for track-driven rice combine harvesters based on stereo vision and details its operational process. Subsequently, we designed a method for acquiring the expected unloading points based on instance segmentation and another method for acquiring the actual unloading points based on geometric location information. Furthermore, this study introduces a method for determining the depth of the grain unloading truck's frame based on the angle of elevation of the unloading tube and proposes a method for acquiring unloading times at each expected point based on a model mapping grain pile height to unloading time. The experimental results confirm the high stability and reliability of the proposed automated unloading system. The maximum relative error between the computed expected unloading points and the actual expected points is less than 4 %. The maximum relative error between the actual unloading points and the true values is also less than 4 %. The mean absolute error between the obtained depth of the grain unloading truck's frame and the actual value is 0.014 m, with a root mean square error of 0.017 m. At each expected unloading point, the mean absolute error between the actual and expected grain pile heights is 0.012 m, with a root mean square error of 0.014 m. Overall, this study effectively enhances the efficiency of agricultural harvesting, reduces labor requirements, and provides a strong impetus for the automation and intelligence of combined harvesting machinery. © 2024

分类号：

  • 相关文献

[1]Deep learning-based software and hardware framework for a noncontact inspection platform for aggregate grading. Jing Qin,Jiabao Wang,Tianjie Lei,Geng Sun,Jianwei Yue,Weiwei Wang,Jinping Chen,Guansheng Qian. 2023

[2]A synthetic datasets based instance segmentation network for High-throughput soybean pods phenotype investigation. Yang, Si,Zheng, Lihua,Yang, Huijun,Zhang, Man,Wu, Tingting,Sun, Shi,Tomasetto, Federico,Wang, Minjuan. 2022

[3]A Query-Based Network for Rural Homestead Extraction from VHR Remote Sensing Images. Ren Wei,Beilei Fan,Yuting Wang,Rongchao Yang. 2023

[4]Instance Segmentation and Number Counting of Grape Berry Images Based on Deep Learning. Yanmin Chen,Xiu Li,Mei Jia,Jiuliang Li,Tianyang Hu,Jun Luo. 2023

[5]Multi-Scale Mixed Attention Tea Shoot Instance Segmentation Model. Chen, Dongmei,Cao, Peipei,Yan, Lijie,Chen, Huidong,Lin, Jia,Li, Xin,Yuan, Lin,Wu, Kaihua. 2024

[6]High-throughput soybean pods high-quality segmentation and seed-per-pod estimation for soybean plant breeding. Yang S.,Zheng L.,Wu T.,Sun S.,Zhang M.,Li M.,Wang M.. 2024

[7]Spatial-channel transformer network based on mask-RCNN for efficient mushroom instance segmentation. Wang, Jiaoling,Song, Weidong,Zheng, Wengang,Feng, Qingchun,Wang, Mingfei,Zhao, Chunjiang. 2024

[8]PodNet: Pod real-time instance segmentation in pre-harvest soybean fields. Shuo Zhou,Qixin Sun,Ning Zhang,Xiujuan Chai,Tan Sun. 2025

作者其他论文 更多>>