Hyperspectral detection system for nitrogen and phosphorus contents of citrus leaves based on cloud edge-to-end architecture

GAO Changlun; ZHANG Fangren; TANG Ting; WU Weibin; DUAN Yuxin; LUO Qing; LIN Huarui; GAO Ting

doi:10.7671/j.issn.1001-411X.202404019

Journal of South China Agricultural University > 2025 > 46(2): 278-286. > DOI: 10.7671/j.issn.1001-411X.202404019

GAO Changlun, ZHANG Fangren, TANG Ting, et al. Hyperspectral detection system for nitrogen and phosphorus contents of citrus leaves based on cloud edge-to-end architecture[J]. Journal of South China Agricultural University, 2025, 46(2): 278-286. DOI: 10.7671/j.issn.1001-411X.202404019

Citation:

PDF (1212 KB)

Hyperspectral detection system for nitrogen and phosphorus contents of citrus leaves based on cloud edge-to-end architecture

1.
College of Engineering, South China Agricultural University, Guangzhou 510642, China
2.
Guangdong E&T Research Center for Mountainous, Orchard Machinery, Guangzhou 510642, China
3.
College of Natural Resources and Environment, South China Agricultural University, Guangzhou 510642, China

More Information

Received Date: April 08, 2024
Available Online: December 30, 2024
Published Date: December 29, 2024

Graphical Abstract

Abstract

Abstract

Objective
To design a hyperspectral detection system for nitrogen and phosphorus contents in citrus leaves.
Method
Based on SR-GRU network training, an inversion model of nitrogen and phosphorus contents was constructed to obtain the spectral data of citrus leaves and corresponding nitrogen and phosphorus contents. A detection system of nitrogen and phosphorus contents in citrus leaves based on cloud edge-to-end architecture was designed. An improved iForest-SAM algorithm was proposed for outlier spectra test and rejection of spectral signals which were easily disturbed by outdoor light. The sparse LoRa message based on over-complete learning dictionary was proposed for fast transmission of spectral data with multiple bands, large size and slow transmission. The edge end of the system was acted as a LoRa gateway in the citrus orchard, and at the mobile terminal end, the sparse LoRa messages were sent to the cloud end via the edge end to load the inversion model for prediction.
Result
The SR-GRU inversion model had the best inversion effect on the contents of nitrogen and phosphorus in citrus leaves, with the determination coefficients of 0.929 and 0.865 respectively, and the normalized root mean square error of 0.083 and 0.079 respectively. The system took less than 1 s to detect nitrogen and phosphorus contents of citrus leaves once, and the LoRa node was connected stably. The web program based on the internet ran stably, and the average page loading time was less than 0.5 s.
Conclusion
The system meets the practical application requirements for timely detection of nitrogen and phosphorus contents in citrus leaves.
- Nitrogen and phosphorus contents,
- Citrus leaves,
- Hyperspectra,
- Deep learning,
- Non-destructive detection,
- Detection system

FullText(HTML)

References (33)

References

[1]	李苹, 付弘婷, 逄玉万, 等. 有机肥配施化肥对柑橘养分吸收及土壤酶活力的影响[J]. 中国土壤与肥料, 2022(3): 39-45.
[2]	LIAO L, FU J L, DONG T T, et al. Effects of nitrogen supply on the photosynthetic capacity of the hybrid citrus cultivar ‘Huangguogan’[J]. Photosynthetica, 2019, 57(2): 581-589.
[3]	WU S W, LI M, ZHANG C M, et al. Effects of phosphorus on fruit soluble sugar and citric acid accumulations in citrus[J]. Plant Physiology and Biochemistry, 2021, 160: 73-81.
[4]	刘钦普. 国内农田氮磷面源污染风险控制研究进展[J]. 江苏农业科学, 2018, 46(1): 1-5.
[5]	KADYAMPAKENI D M, MORGAN K T, NKEDI-KIZZA P, et al. Nutrient management options for Florida citrus: A review of NPK application and analytical methods[J]. Journal of Plant Nutrition, 2015, 38(4): 568-583. doi: 10.1080/01904167.2014.934470
[6]	YANG X C, LEI S G, ZHAO Y B, et al. Use of hyperspectral imagery to detect affected vegetation and heavy metal polluted areas: A coal mining area, China[J]. Geocarto International, 2022, 37(10): 2893-2912.
[7]	罗锡文, 廖娟, 臧英, 等. 我国农业生产的发展方向: 从机械化到智慧化[J]. 中国工程科学, 2022, 24(6): 46-54.
[8]	ARIEA RAMIREZ W, MISHRA G, PANDA B K, et al. Multispectral camera system design for replacement of hyperspectral cameras for detection of aflatoxin B₁[J]. Computers and Electronics in Agriculture, 2022, 198: 107078. doi: 10.1016/j.compag.2022.107078.
[9]	THIEN PHAM Q, LIOU N S. The development of on-line surface defect detection system for jujubes based on hyperspectral images[J]. Computers and Electronics in Agriculture, 2022, 194: 106743. doi: 10.1016/j.compag.2022.106743.
[10]	GUO W, LI X X, XIE T H. Method and system for nondestructive detection of freshness in Penaeus vannamei based on hyperspectral technology[J]. Aquaculture, 2021, 538: 736512. doi: 10.1016/j.aguaculture.2021.736512.
[11]	刘翠玲, 闻世震, 孙晓荣, 等. 基于云计算的食品品质实时在线光谱检测系统实现[J]. 食品科学技术学报, 2023, 41(2): 161-170.
[12]	王楠, 李震, 李佳盟, 等. 融合多光谱成像与深度学习的作物植株叶绿素检测系统研究[J]. 农业机械学报, 2023, 54(3): 260-269.
[13]	蔡健荣, 黄楚钧, 马立鑫, 等. 一维卷积神经网络的手持式可见/近红外柑橘可溶性固形物含量无损检测系统[J]. 光谱学与光谱分析, 2023, 43(4): 2792-2798.
[14]	COSTA L, KUNWAR S, AMPATZIDIS Y, et al. Determining leaf nutrient concentrations in citrus trees using UAV imagery and machine learning[J]. Precision Agriculture, 2022, 23(3): 854-875.
[15]	OSCO L P, RAMOS A P M, FAITA PINHEIRO M M, et al. A machine learning framework to predict nutrient content in Valencia-orange leaf hyperspectral measurements[J]. Remote Sensing, 2020, 12(6): 906. doi: 10.3390/rs12060906.
[16]	岳学军, 凌康杰, 王林惠, 等. 基于高光谱和深度迁移学习的柑橘叶片钾含量反演[J]. 农业机械学报, 2019, 50(6): 186-195. doi: 10.6041/j.issn.1000-1298.2019.03.020
[17]	代秋芳, 廖臣龙, 李震, 等. 基于CARS-CNN的高光谱柑橘叶片含水率可视化研究[J]. 光谱学与光谱分析, 2022, 42(9): 2848-2854. doi: 10.3964/j.issn.1000-0593(2022)09-2848-07
[18]	SU H J, JIA C L, ZHENG P, et al. Superpixel-based weighted collaborative sparse regression and reweighted low-rank representation for hyperspectral image unmixing[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 393-408.
[19]	PENG G. Learning the sparse prior: Modern approaches[J]. Wiley Interdisciplinary Reviews-Computational Statistics, 2024, 16(1): 1646. doi: 10.1002/wics.1646.
[20]	ZHANG X W, XIE L J, WANG J P. Some results on OMP algorithm for MMV problem[J]. Mathematical Methods in the Applied Sciences, 2022, 45(9): 5402-5411.
[21]	ZOU H B, WU S S, TIAN M X. Radar quantitative precipitation estimation based on the gated recurrent unit neural network and echo-top data[J]. Advances in Atmospheric Sciences, 2023, 40(6): 1043-1057.
[22]	李亚茹, 张宇来, 王佳晨. 面向超参数估计的贝叶斯优化方法综述[J]. 计算机科学, 2022, 49(3): 86-92.
[23]	LIU W W, LI M J, ZHANG M Y, et al. Hyperspectral inversion of mercury in reed leaves under different levels of soil mercury contamination[J]. Environmental Science and Pollution Research, 2020, 27(18): 22935-22945.
[24]	PENG Y, ZHANG M, XU Z Y, et al. Estimation of leaf nutrition status in degraded vegetation based on field survey and hyperspectral data[J]. Scientific Reports, 2020, 10: 4361. doi: 10.1038/s41598-020-61294-7.
[25]	黄承伟, 戴连奎, 董学锋. 结合SNV的分段直接标准化方法在拉曼光谱模型传递中的应用[J]. 光谱学与光谱分析, 2011, 31(9): 1279-1282. doi: 10.3964/j.issn.1000-0593(2011)05-1279-04
[26]	YOON H I, LEE H, YANG J, et al. Predicting models for plant metabolites based on PLSR, AdaBoost, XGBoost, and LightGBM algorithms using hyperspectral imaging of Brassica juncea[J]. Agriculture-Basel, 2023, 13(8): 1477. doi: 10.3990/agriculture13081477.
[27]	SHENG H, CHI H X, XU M M, et al. Inland water chemical oxygen demand estimation based on improved SVR for hyperspectral data[J]. Spectroscopy and Spectral Analysis, 2021, 41(10): 3565-3571.
[28]	XUE Y C, ZHU C Y, JIANG H. Comparison of the performance of different one-dimensional convolutional neural network models-based near-infrared spectra for determination of chlorpyrifos residues in corn oil[J]. Infrared Physics & Technology, 2023, 132: 104734. doi: 10.1016/j.infrared.2024.104734.
[29]	WANG Y, HE Y H, WANG Z G, et al. Information fusion technology for terahertz spectra and hyperspectral imaging in wood species identification[J]. European Journal of Wood and Wood Products, 2024, 82(3): 579-589.
[30]	SÁDECKÁ J, JAKUBÍKOVÁ M. Varietal classification of white wines by fluorescence spectroscopy[J]. Journal of Food Science and Technology-Mysore, 2020, 57(7): 2545-2553.
[31]	YANG C B, XU J, FENG M C, et al. Evaluation of hyperspectral monitoring model for aboveground dry biomass of winter wheat by using multiple factors[J]. Agronomy-Basel, 2023, 13(4): 983. doi: 10.3390/agronomy13040983.
[32]	张哲宇, 李耀翔, 王志远, 等. 基于IFSR异常样本剔除的落叶松木材密度近红外优化模型的研究[J]. 光谱学与光谱分析, 2022, 42(11): 3395-3402.
[33]	刘煊, 渠慎明. 低秩稀疏和改进SAM的高光谱图像误标签检测[J]. 激光技术, 2022, 46(6): 808-816.

Cited By

Get Citation

PDF

XML

Article views (505) PDF downloads (44)

Turn off MathJax

Article Contents

Abstract

References

Hyperspectral detection system for nitrogen and phosphorus contents of citrus leaves based on cloud edge-to-end architecture

Abstract

References

Catalog

Export File

Citation

Format

Content