数字农科院2.0

Detection method of broken grains and impurities in harvested soybeans using feature wavelength selection and MobileNetV4-Unet-SGCPNet hybrid network

文献类型： 外文期刊

作者： Jin, Chengqian;Cheng, Gong;Chang, Zhichang;Chen, Man;Yang, Tengxiang;Shi, Yinyan;Gai, Xiaobin

作者机构：

关键词： Soybeans;Breakage;Foreign matter content;Spectral detection;Classification and identification

期刊名称： SMART AGRICULTURAL TECHNOLOGY

ISSN：

年卷期： 2025 年 13 卷

页码：

收录情况： ESCI（2025版）

摘要： To address the challenges of hyperspectral data redundancy, small-target segmentation difficulty, and insufficient model real-time performance in high-precision online detection of foreign matter and kernel breakage in machine-harvested soybeans, this study proposes a collaborative detection method based on feature wavelength optimization + MobileNetV4-Unet-SGCPNet hybrid network. First, 18 key feature bands were screened from 400 - 1000 nm hyperspectral data using successive projection algorithm (SPA) and competitive adaptive reweighted sampling (CARS), constructing a multi-source feature spectral image dataset. Subsequently, a MobileNetV4-Unet-SGCPNet hybrid network was designed, with a lightweight MobileNetV4 as the encoder, combined with the symmetric encoder-decoder structure of Unet and the spatial detail-guided context propagation module (SGCP) to achieve high-precision segmentation of broken grains, complete grains, and impurities. Finally, pixel-wise voting was employed to fuse multi-band feature information, enhancing the model's generalization capability. The results demonstrate that: on the test set, the model achieves an average intersection over - union of 89.69 % for soybean component recognition, with a mean precision average of 94.55 %, a mean precision of 93.63 %, a frame rate of 4.77 FPS, a parameter count of only 2.86 MB, and a computational load of 35.78 GFLOPs. Compared to mainstream models, this method reduces parameters by 97.2 % and computational cost by 97.8 %, while the average intersection - over - union drops by only 6.2 %, with a frame rate improvement of over 5 times, striking a significant balance between detection accuracy and real-time performance. Crossvariety and cross-device validations further confirm that the model effectively adapts to morphological and spectral variations across different soybean varieties, exhibiting strong generalization ability. This study provides a core algorithmic foundation for online monitoring systems of intelligent harvester operation quality, offering critical support for enhancing the commercial value of machine-harvested soybeans and advancing the intelligence level of agricultural machinery.

分类号：

  • 相关文献

[1]Determination and parameters calibration of the soybean-bonded particle model based on discrete element method. Han, Dan-Dan,Wang, Qing,Wang, Yun-Xia,Li, Wei,Tang, Chao,Lv, Xiao-Rong. 2024

[2]A Spectral Device for Rice Chlorophyll Content Detection with Background Classification Capability. Chen, Yanyu,Wang, Xiaochan,Xing, Xiaoyang,Xu, Sheng,Zhang, Xiaolei,Chen, Shengfeng,Yin, Yue,Wang, Dezhi,Xi, Rui,Xu, Xin. 2026

[3]Online Monitoring Method of Mechanized Soybean Harvest Quality Based on Machine Vision [基于机器视觉的大豆机械化收获质量在线监测方法]. Chen M.,Ni Y.,Jin C.,Xu J.,Zhang G.. 2021

[4]Classification and Recognition of Soybean Quality Based on Hyperspectral Imaging and Random Forest Methods. Man Chen,Zhichang Chang,Chengqian Jin,Gong Cheng,Shiguo Wang,Youliang Ni. 2025

[5]Salt Tolerance in Soybean. Phang, Tsui-Hung,Shao, Guihua,Lam, Hon-Ming,Shao, Guihua,Phang, Tsui-Hung,Lam, Hon-Ming.

[6]Induction of phytochemical glyceollins accumulation in soybean following treatment with biotic elicitor (Aspergillus oryzae). Eromosele, Ojokoh,Bo, Shi,Ping, Liang. 2013

[7]Effect of parasitism on flight behavior of the soybean aphid, Aphis glycines. Wu, Kongming,Wyckhuys, Kris A. G.,Heimpel, George E..

[8]Soybean crushing forms by mechanical harvesting and factors affecting the proportions of different forms. Youliang Ni,Chengqian Jin,Man Chen,Zhenjie Qian,Tengxiang Yang,Jinshan Xu,Gangwei Liu. 2023

[9]An Empirical Analysis of Global Soybean Supply Potential and China’s Diversified Import Strategies Based on Global Agro-Ecological Zones and Multi-Objective Nonlinear Programming Models. Xiaoyu Jiang,Huishang Li,Xin Dai,Jingdong Li,Ye Liu. 2025

[10]A Review of Key Technologies and Intelligent Applications in Soybean Mechanized Harvesting: Chinese and International Perspectives. Wang, Jin,Shan, Chunli,Gou, Fuqiang,Qian, Zhenjie,Ni, Youliang,Liu, Zheng,Jin, Chengqian. 2025

作者其他论文 更多>>