DROUGHT RECOGNITION FOR THE SOYBEAN PLANT BASED ON LIGHTWEIGHT DEEP LEARNING MODEL
Keywords:
Soybean drought, Multispectral image, Feature screening, Recognition model, Mobile appAbstract
To address the issues of low soybean self-sufficiency and drought - related production constraints, multispectral imaging can non - invasively detect crop characteristics, while deep convolutional networks identify drought from multispectral images. However, large models face mobility limitations due to high computational requirements.This study presents a mobile detection approach integrating ReliefF feature screening and lightweight convolutional neural networks, and develops the "Early Acknowledgment for Soybean Drought" App. It embeds a lightweight model that optimizes 37 - dimensional soybean canopy multispectral features via ReliefF and uses a three - layer 1D convolutional network for drought identification.The model achieves 96.88% classification accuracy on the self - built dataset, with an inference time of 18 ms, a size under 30 MB, and less than 60 MB memory usage on mobile. The APP integrates the multispectral camera SDK and PyTorch inference engine, enabling real - time spectral analysis. Field tests show its one - button operation, low learning curve for farmers, and significant water - saving and yield - increasing effects, offering a lightweight, high - precision mobile solution for soybean drought management and promoting smart agriculture development.References
[1] Li Hefeng. Steady increase of domestic soybean production capacity and self-sufficiency. Economic Daily News, 2024-11-22(006). DOI: 10.28425/n.cnki.njjrb.2024.008771.
[2] Shen Panpan. Research on the calculation method of soybean canopy wilt index based on multispectral image processing. Heilongjiang Bayi Nongken University, 2023. DOI:10.27122/d.cnki.ghlnu.2023.000122.
[3] Haoxin Song. Research on citrus information extraction from UAV multispectral images based on deep learning. Guilin University of Technology, 2023. DOI: 10.27050/d.cnki.ghlgc.2023.001051.
[4] Gao Shijiao, Guan Haiou, Ma Xiaodan, et al. A multispectral image extraction method for soybean canopy. Spectroscopy and Spectral Analysis, 2022, 42(11): 3568-3574.
[5] Fu Hongyu, Wang Wei, Lu Jianning, et al. Estimation of physical and chemical traits of ramie based on multi-spectral remote sensing by unmanned aircraft and machine learning. Journal of Agricultural Machinery, 2023, 54(05): 194-200+347.
[6] Fan Xuexing, Zhang Huichun, Zou Yiping, et al. Inversion of plant chlorophyll content based on multispectral imaging and machine learning. Forestry Science, 2023, 59(07): 78-88.
[7] Han Wenting, Cui Jiawei, Cui Xin, et al. Research on salinity estimation of agricultural soil based on feature optimization and machine learning. Journal of Agricultural Machinery, 2023, 54(03): 328-337.
[8] Huang Linsheng, Shao Song, Lu Xianju, et al. Multispectral image segmentation and alignment of lettuce based on convolutional neural network. Journal of Agricultural Machinery, 2021, 52(09): 186-194.
[9] Pang Q. Research on the detection of obvious/hidden defects in apple epidermis based on deep learning and spectral imaging . Shanghai Ocean University, 2022. DOI:10.27314/d.cnki.gsscu.2022.000147.
[10] Bedi S R, Singh A. A Feature Selection Based Relief Algorithm with Fuzzy Logic for Software Effort Estimation.Research Cell: An International Journal of Engineering Sciences, 2018, 30(SP): 1-5.
[11] Jiang C, Sun X, Dai Y, et al. EEG Emotion Recognition Employing RGPCN-BiGRUAM: ReliefF-Based Graph Pooling Convolutional Network and BiGRU Attention Mechanism. Electronics, 2024, 13(13): 2530-2530.
[12] Mahbod A, Saeidi N, Hatamikia S, et al. Evaluating pre-trained convolutional neural networks and foundation models as feature extractors for content-based medical image retrieval. Engineering Applications of Artificial Intelligence, 2025: 150110571-110571.
[13] Yifan D, Chuanbo W, Zheng W. A motor bearing fault diagnosis method based on multi-source data and onedimensional lightweight convolution neural network. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, 2023, 237(2): 272-283.
[14] Carmona A M ,Sautua J F, Pérez-Hernández O, et al. AgroDecisor EFC: First Android TM app decision support tool for timing fungicide applications for management of late-season soybean diseases. Computers and Electronics in Agriculture, 2018: 144310-313.