A comprehensive benchmark and dual-encoder baseline for terraced field extraction in complex mountainous regions

Terraces play a vital role in mitigating runoff and intercepting sediment. Large-scale, high-precision extraction of terraces from remote sensing imagery is essential for environmental monitoring and soil erosion control. However, accurate spatial identification of terraces still presents significant challenges, including scale variability, shadow interference, and complex geometric configurations. To address these issues, we propose DETS-Net, a dual-encoder network that integrates CNN and Transformer architectures. The network employs ResNet-34 as the CNN encoder for local feature detail extraction, utilizes a Transformer encoder to strengthen long-range dependency modeling, and incorporates an adaptive feature fusion (AFF) module for the dynamic integration of the two feature branches. We develop the THDIR dataset, which covers four typical terrain types, namely, plateau mountains, hills, and mountain-basin transition zones, and contains 9,336 standardized 256 × 256 pixel images. We conduct systematic comparisons with eight representative deep learning models. Experimental results demonstrate that DETS-Net achieves an overall accuracy (OA) of 89.91%, a mean intersection over union (MIoU) of 81.28%, and an F1-score of 87.99% on the THDIR dataset, outperforming current mainstream methods in segmentation precision. We apply the optimized model to 2023 remote sensing imagery of Luotian County in the Dabie Mountains, Hubei Province, and generate a high-precision county-level terrace spatial distribution map. This framework supports the assessment of soil and water conservation benefits through area statistics, slope distribution data, and high-precision boundary extraction and provides a reproducible technical pathway for automated terrace mapping using high-resolution remote sensing imagery in complex mountainous regions.

Authors

Institutions

Publication Details

Journal
Frontiers in Ecology and Evolution
Published
2026-11-13
DOI
https://doi.org/10.3389/fevo.2026.2011654
Primary Topic
Remote-Sensing Image Classification
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A comprehensive benchmark and dual-encoder baseline for terraced field extraction in complex mountainous regions

Ruoxuan Deng, Pei Tian, Yuanzhe Cai, Yan Chen et al.
5 links
Frontiers in Ecology and Evolution
Remote-Sensing Image Classification
article

A comprehensive benchmark and dual-encoder baseline for terraced field extraction in complex mountainous regions

Ruoxuan Deng, Pei Tian, Yuanzhe Cai, Yan Chen, Biao Ma, Ruihan Xiao, Yujie Sun, Zheng Zhang
article en

Abstract

Terraces play a vital role in mitigating runoff and intercepting sediment. Large-scale, high-precision extraction of terraces from remote sensing imagery is essential for environmental monitoring and soil erosion control. However, accurate spatial identification of terraces still presents significant challenges, including scale variability, shadow interference, and complex geometric configurations. To address these issues, we propose DETS-Net, a dual-encoder network that integrates CNN and Transformer architectures. The network employs ResNet-34 as the CNN encoder for local feature detail extraction, utilizes a Transformer encoder to strengthen long-range dependency modeling, and incorporates an adaptive feature fusion (AFF) module for the dynamic integration of the two feature branches. We develop the THDIR dataset, which covers four typical terrain types, namely, plateau mountains, hills, and mountain-basin transition zones, and contains 9,336 standardized 256 × 256 pixel images. We conduct systematic comparisons with eight representative deep learning models. Experimental results demonstrate that DETS-Net achieves an overall accuracy (OA) of 89.91%, a mean intersection over union (MIoU) of 81.28%, and an F1-score of 87.99% on the THDIR dataset, outperforming current mainstream methods in segmentation precision. We apply the optimized model to 2023 remote sensing imagery of Luotian County in the Dabie Mountains, Hubei Province, and generate a high-precision county-level terrace spatial distribution map. This framework supports the assessment of soil and water conservation benefits through area statistics, slope distribution data, and high-precision boundary extraction and provides a reproducible technical pathway for automated terrace mapping using high-resolution remote sensing imagery in complex mountainous regions.

Frontiers in Ecology and EvolutionVol. 14
Central China Normal University (CN), Changjiang Institute of Survey, Planning, Design and Research (CN), Shenzhen Technology University (CN)
Remote-Sensing Image Classification
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon