TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection

Accurate detection of soil total nitrogen (TN) is crucial for enhancing crop growth and quality. However, soil variability across different regions limits the generalization ability of calibration models. To address this challenge, this study introduces a model named transfer-learning-assisted laser...

Full description

Saved in:
Bibliographic Details
Main Authors: Peng Lin, Shixiang Ma, Zhizheng Shi, Peiying Li, Leizi Jiao, Hongwu Tian, Zhen Xing, Chunjiang Zhao, Daming Dong
Format: Article
Language:English
Published: Elsevier 2025-12-01
Series:Smart Agricultural Technology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S277237552500351X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849424618422534144
author Peng Lin
Shixiang Ma
Zhizheng Shi
Peiying Li
Leizi Jiao
Hongwu Tian
Zhen Xing
Chunjiang Zhao
Daming Dong
author_facet Peng Lin
Shixiang Ma
Zhizheng Shi
Peiying Li
Leizi Jiao
Hongwu Tian
Zhen Xing
Chunjiang Zhao
Daming Dong
author_sort Peng Lin
collection DOAJ
description Accurate detection of soil total nitrogen (TN) is crucial for enhancing crop growth and quality. However, soil variability across different regions limits the generalization ability of calibration models. To address this challenge, this study introduces a model named transfer-learning-assisted laser-induced breakdown spectroscopy for cross-regional soil analysis (TransLIBS-CRS), specifically designed to mitigate the low cross-domain prediction accuracy caused by variations in regional soil properties. By fine-tuning with a limited number of target domain samples, this method significantly improves the cross-domain applicability of LIBS data, alleviating the challenges associated with the difficulty of obtaining soil samples from diverse regions. In the task of predicting TN in Guangzhou using the Beijing dataset, the TransLIBS-CRS model achieved optimal performance, with RV2 of 0.846 and RMSEV- of 0.814 g/kg. Further analysis through saliency map and chemometric methods revealed that spectral lines of carbon at 193.0 nm and 247.8 nm play a key role in the quantitative detection of TN. Notably, these spectral features also demonstrated stable predictive contributions when transferred to the Guangzhou soil dataset. This approach offers a feasible solution for large-scale and efficient soil TN detection.
format Article
id doaj-art-9cecc27666f748d2913e9dfb0b9f27b4
institution Kabale University
issn 2772-3755
language English
publishDate 2025-12-01
publisher Elsevier
record_format Article
series Smart Agricultural Technology
spelling doaj-art-9cecc27666f748d2913e9dfb0b9f27b42025-08-20T03:30:04ZengElsevierSmart Agricultural Technology2772-37552025-12-011210111810.1016/j.atech.2025.101118TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detectionPeng Lin0Shixiang Ma1Zhizheng Shi2Peiying Li3Leizi Jiao4Hongwu Tian5Zhen Xing6Chunjiang Zhao7Daming Dong8College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China; Research Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaResearch Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, China; Corresponding authors.Research Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaCollege of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China; Research Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaResearch Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaResearch Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaResearch Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaCollege of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China; Research Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, ChinaResearch Center of Intelligent Equipment, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China; Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Beijing 100097, China; Corresponding authors.Accurate detection of soil total nitrogen (TN) is crucial for enhancing crop growth and quality. However, soil variability across different regions limits the generalization ability of calibration models. To address this challenge, this study introduces a model named transfer-learning-assisted laser-induced breakdown spectroscopy for cross-regional soil analysis (TransLIBS-CRS), specifically designed to mitigate the low cross-domain prediction accuracy caused by variations in regional soil properties. By fine-tuning with a limited number of target domain samples, this method significantly improves the cross-domain applicability of LIBS data, alleviating the challenges associated with the difficulty of obtaining soil samples from diverse regions. In the task of predicting TN in Guangzhou using the Beijing dataset, the TransLIBS-CRS model achieved optimal performance, with RV2 of 0.846 and RMSEV- of 0.814 g/kg. Further analysis through saliency map and chemometric methods revealed that spectral lines of carbon at 193.0 nm and 247.8 nm play a key role in the quantitative detection of TN. Notably, these spectral features also demonstrated stable predictive contributions when transferred to the Guangzhou soil dataset. This approach offers a feasible solution for large-scale and efficient soil TN detection.http://www.sciencedirect.com/science/article/pii/S277237552500351XLaser-induced breakdown spectroscopyTransfer learningSoil total nitrogenDifferent regionsExplainability
spellingShingle Peng Lin
Shixiang Ma
Zhizheng Shi
Peiying Li
Leizi Jiao
Hongwu Tian
Zhen Xing
Chunjiang Zhao
Daming Dong
TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection
Smart Agricultural Technology
Laser-induced breakdown spectroscopy
Transfer learning
Soil total nitrogen
Different regions
Explainability
title TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection
title_full TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection
title_fullStr TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection
title_full_unstemmed TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection
title_short TransLIBS-CRS: Integrating LIBS with transfer learning for accurate cross-regional soil total nitrogen detection
title_sort translibs crs integrating libs with transfer learning for accurate cross regional soil total nitrogen detection
topic Laser-induced breakdown spectroscopy
Transfer learning
Soil total nitrogen
Different regions
Explainability
url http://www.sciencedirect.com/science/article/pii/S277237552500351X
work_keys_str_mv AT penglin translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT shixiangma translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT zhizhengshi translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT peiyingli translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT leizijiao translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT hongwutian translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT zhenxing translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT chunjiangzhao translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection
AT damingdong translibscrsintegratinglibswithtransferlearningforaccuratecrossregionalsoiltotalnitrogendetection