Mapping of soil sampling sites using terrain and hydrological attributes

Efficient soil sampling is essential for effective soil management and research on soil health. Traditional site selection methods are labor-intensive and fail to capture soil variability comprehensively. This study introduces a deep learning-based tool that automates soil sampling site selection us...

Full description

Saved in:
Bibliographic Details
Main Authors: Tan-Hanh Pham, Kristopher Osterloh, Kim-Doang Nguyen
Format: Article
Language:English
Published: KeAi Communications Co., Ltd. 2025-09-01
Series:Artificial Intelligence in Agriculture
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2589721725000479
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849387480511414272
author Tan-Hanh Pham
Kristopher Osterloh
Kim-Doang Nguyen
author_facet Tan-Hanh Pham
Kristopher Osterloh
Kim-Doang Nguyen
author_sort Tan-Hanh Pham
collection DOAJ
description Efficient soil sampling is essential for effective soil management and research on soil health. Traditional site selection methods are labor-intensive and fail to capture soil variability comprehensively. This study introduces a deep learning-based tool that automates soil sampling site selection using spectral images. The proposed framework consists of two key components: an extractor and a predictor. The extractor, based on a convolutional neural network (CNN), derives features from spectral images, while the predictor employs self-attention mechanisms to assess feature importance and generate prediction maps. The model is designed to process multiple spectral images and address the class imbalance in soil segmentation.The model was trained on a soil dataset from 20 fields in eastern South Dakota, collected via drone-mounted LiDAR with high-precision GPS. Evaluation on a test set achieved a mean intersection over union (mIoU) of 69.46 % and a mean Dice coefficient (mDc) of 80.35 %, demonstrating strong segmentation performance. The results highlight the model's effectiveness in automating soil sampling site selection, providing an advanced tool for producers and soil scientists. Compared to existing state-of-the-art methods, the proposed approach improves accuracy and efficiency, optimizing soil sampling processes and enhancing soil research.
format Article
id doaj-art-5726a9ff93914f79bd601fcb0aed852e
institution Kabale University
issn 2589-7217
language English
publishDate 2025-09-01
publisher KeAi Communications Co., Ltd.
record_format Article
series Artificial Intelligence in Agriculture
spelling doaj-art-5726a9ff93914f79bd601fcb0aed852e2025-08-20T03:53:47ZengKeAi Communications Co., Ltd.Artificial Intelligence in Agriculture2589-72172025-09-0115347048110.1016/j.aiia.2025.04.007Mapping of soil sampling sites using terrain and hydrological attributesTan-Hanh Pham0Kristopher Osterloh1Kim-Doang Nguyen2Department of Mechanical and Civil Engineering, Florida Institute of Technology, USADepartment of Agronomy, Horticulture & Plant Science, South Dakota State University, USADepartment of Mechanical and Civil Engineering, Florida Institute of Technology, USA; Corresponding author.Efficient soil sampling is essential for effective soil management and research on soil health. Traditional site selection methods are labor-intensive and fail to capture soil variability comprehensively. This study introduces a deep learning-based tool that automates soil sampling site selection using spectral images. The proposed framework consists of two key components: an extractor and a predictor. The extractor, based on a convolutional neural network (CNN), derives features from spectral images, while the predictor employs self-attention mechanisms to assess feature importance and generate prediction maps. The model is designed to process multiple spectral images and address the class imbalance in soil segmentation.The model was trained on a soil dataset from 20 fields in eastern South Dakota, collected via drone-mounted LiDAR with high-precision GPS. Evaluation on a test set achieved a mean intersection over union (mIoU) of 69.46 % and a mean Dice coefficient (mDc) of 80.35 %, demonstrating strong segmentation performance. The results highlight the model's effectiveness in automating soil sampling site selection, providing an advanced tool for producers and soil scientists. Compared to existing state-of-the-art methods, the proposed approach improves accuracy and efficiency, optimizing soil sampling processes and enhancing soil research.http://www.sciencedirect.com/science/article/pii/S2589721725000479Agriculture precisionDeep learningSoil samplingSpectral imagingSegmentation
spellingShingle Tan-Hanh Pham
Kristopher Osterloh
Kim-Doang Nguyen
Mapping of soil sampling sites using terrain and hydrological attributes
Artificial Intelligence in Agriculture
Agriculture precision
Deep learning
Soil sampling
Spectral imaging
Segmentation
title Mapping of soil sampling sites using terrain and hydrological attributes
title_full Mapping of soil sampling sites using terrain and hydrological attributes
title_fullStr Mapping of soil sampling sites using terrain and hydrological attributes
title_full_unstemmed Mapping of soil sampling sites using terrain and hydrological attributes
title_short Mapping of soil sampling sites using terrain and hydrological attributes
title_sort mapping of soil sampling sites using terrain and hydrological attributes
topic Agriculture precision
Deep learning
Soil sampling
Spectral imaging
Segmentation
url http://www.sciencedirect.com/science/article/pii/S2589721725000479
work_keys_str_mv AT tanhanhpham mappingofsoilsamplingsitesusingterrainandhydrologicalattributes
AT kristopherosterloh mappingofsoilsamplingsitesusingterrainandhydrologicalattributes
AT kimdoangnguyen mappingofsoilsamplingsitesusingterrainandhydrologicalattributes