SOIL TOTAL NITROGEN SPECTRAL INVERSION BASED ON THE DANN MODEL

Zhenyuan Ji; Ao Yang; Zhiguo Liang; Haoxuan Yuan; Yun Zhang; Yong Du; Yaohua Li

doi:10.1109/IGARSS55030.2025.11242272

SOIL TOTAL NITROGEN SPECTRAL INVERSION BASED ON THE DANN MODEL

Zhenyuan Ji
, Ao Yang
, Zhiguo Liang
, Haoxuan Yuan^*
, Yun Zhang
, Yong Du
, Yaohua Li

^*Corresponding author for this work

School of Electronics and Information Engineering

Harbin Institute of Technology
China Mobile Chengdu Institute of Research and Development

Research output: Contribution to journal › Conference article › peer-review

Abstract

Soil degradation threatens agricultural sustainability and food security. This study explores the effects of spectral preprocessing methods and machine learning models on predicting soil total nitrogen using the LUCAS2009 dataset. Preprocessing techniques, including first derivative (FD) and detrending (DT), enhanced spectral sensitivity. Machine learning models such as partial least squares regression (PLSR), support vector machines (SVM), random forest (RF), and a novel Density-Aware Neural Network (DANN) were evaluated. DANN achieved the best performance on FD data, with R² = 0.872, RMSE = 1.362, and RPD = 2.795. These findings provide insights into optimizing soil spectral analysis for precision agriculture.

Original language	English
Pages (from-to)	4363-4366
Number of pages	4
Journal	International Geoscience and Remote Sensing Symposium (IGARSS)
DOIs	https://doi.org/10.1109/IGARSS55030.2025.11242272
State	Published - 2025
Event	2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025 - Brisbane, Australia Duration: 3 Aug 2025 → 8 Aug 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 2 Zero Hunger

Keywords

DANN
Deep learning
soil total nitrogen
spectral preprocessing

Access to Document

10.1109/IGARSS55030.2025.11242272

Cite this

@article{942703c4b3fb4dbb998e914921ca8308,

title = "SOIL TOTAL NITROGEN SPECTRAL INVERSION BASED ON THE DANN MODEL",

abstract = "Soil degradation threatens agricultural sustainability and food security. This study explores the effects of spectral preprocessing methods and machine learning models on predicting soil total nitrogen using the LUCAS2009 dataset. Preprocessing techniques, including first derivative (FD) and detrending (DT), enhanced spectral sensitivity. Machine learning models such as partial least squares regression (PLSR), support vector machines (SVM), random forest (RF), and a novel Density-Aware Neural Network (DANN) were evaluated. DANN achieved the best performance on FD data, with R2 = 0.872, RMSE = 1.362, and RPD = 2.795. These findings provide insights into optimizing soil spectral analysis for precision agriculture.",

keywords = "DANN, Deep learning, soil total nitrogen, spectral preprocessing",

author = "Zhenyuan Ji and Ao Yang and Zhiguo Liang and Haoxuan Yuan and Yun Zhang and Yong Du and Yaohua Li",

note = "Publisher Copyright: {\textcopyright}2025 IEEE.; 2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025 ; Conference date: 03-08-2025 Through 08-08-2025",

year = "2025",

doi = "10.1109/IGARSS55030.2025.11242272",

language = "英语",

pages = "4363--4366",

journal = "International Geoscience and Remote Sensing Symposium (IGARSS)",

issn = "2153-6996",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - SOIL TOTAL NITROGEN SPECTRAL INVERSION BASED ON THE DANN MODEL

AU - Ji, Zhenyuan

AU - Yang, Ao

AU - Liang, Zhiguo

AU - Yuan, Haoxuan

AU - Zhang, Yun

AU - Du, Yong

AU - Li, Yaohua

PY - 2025

Y1 - 2025

N2 - Soil degradation threatens agricultural sustainability and food security. This study explores the effects of spectral preprocessing methods and machine learning models on predicting soil total nitrogen using the LUCAS2009 dataset. Preprocessing techniques, including first derivative (FD) and detrending (DT), enhanced spectral sensitivity. Machine learning models such as partial least squares regression (PLSR), support vector machines (SVM), random forest (RF), and a novel Density-Aware Neural Network (DANN) were evaluated. DANN achieved the best performance on FD data, with R2 = 0.872, RMSE = 1.362, and RPD = 2.795. These findings provide insights into optimizing soil spectral analysis for precision agriculture.

AB - Soil degradation threatens agricultural sustainability and food security. This study explores the effects of spectral preprocessing methods and machine learning models on predicting soil total nitrogen using the LUCAS2009 dataset. Preprocessing techniques, including first derivative (FD) and detrending (DT), enhanced spectral sensitivity. Machine learning models such as partial least squares regression (PLSR), support vector machines (SVM), random forest (RF), and a novel Density-Aware Neural Network (DANN) were evaluated. DANN achieved the best performance on FD data, with R2 = 0.872, RMSE = 1.362, and RPD = 2.795. These findings provide insights into optimizing soil spectral analysis for precision agriculture.

KW - DANN

KW - Deep learning

KW - soil total nitrogen

KW - spectral preprocessing

UR - https://www.scopus.com/pages/publications/105033580958

U2 - 10.1109/IGARSS55030.2025.11242272

DO - 10.1109/IGARSS55030.2025.11242272

M3 - 会议文章

AN - SCOPUS:105033580958

SN - 2153-6996

SP - 4363

EP - 4366

JO - International Geoscience and Remote Sensing Symposium (IGARSS)

JF - International Geoscience and Remote Sensing Symposium (IGARSS)

T2 - 2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025

Y2 - 3 August 2025 through 8 August 2025

ER -

SOIL TOTAL NITROGEN SPECTRAL INVERSION BASED ON THE DANN MODEL

Abstract

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this