PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model

Saddle-point extraction is essential for accurately identifying topographic features and landforms and conducting geomorphological mapping. However, the widely used positive–negative terrain method (PNTM) is often plagued by a substantial number of false saddle points, a prevalent issue in many extr...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhe Zhou, Hao Wu, Zhenyu Zhang, Bo Kong, Min Yang, Tinghua Ai, Huafei Yu
Format: Article
Language:English
Published: Taylor & Francis Group 2025-08-01
Series:International Journal of Digital Earth
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/17538947.2025.2545583
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849224264119484416
author Zhe Zhou
Hao Wu
Zhenyu Zhang
Bo Kong
Min Yang
Tinghua Ai
Huafei Yu
author_facet Zhe Zhou
Hao Wu
Zhenyu Zhang
Bo Kong
Min Yang
Tinghua Ai
Huafei Yu
author_sort Zhe Zhou
collection DOAJ
description Saddle-point extraction is essential for accurately identifying topographic features and landforms and conducting geomorphological mapping. However, the widely used positive–negative terrain method (PNTM) is often plagued by a substantial number of false saddle points, a prevalent issue in many extraction techniques. To address this challenge, this study presents a novel model that combines the PNTM with a convolutional neural network (CNN) called PNTM-CNN. In this approach, candidate saddle points are first identified using the PNTM and then refined using a CNN that integrates multiscale topographic features. The experimental results indicate that the PNTM-CNN model, which leverages four scales of features (elevation, aspect, curvature, slope, and hillshade), effectively reduces the occurrence of false saddle points, achieving a precision of 89%, a recall of 83%, and an F1 score of 85%. This performance significantly exceeds that of the traditional moving window analysis and topological association methods. Although the automation level of the PNTM-CNN model requires improvement, the integration of deep learning methods offers new insights for addressing complex topographic feature extraction challenges and shows a promising application potential.
format Article
id doaj-art-0219c19f79eb43bf8a8fe5463edcf862
institution Kabale University
issn 1753-8947
1753-8955
language English
publishDate 2025-08-01
publisher Taylor & Francis Group
record_format Article
series International Journal of Digital Earth
spelling doaj-art-0219c19f79eb43bf8a8fe5463edcf8622025-08-25T11:28:18ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552025-08-0118110.1080/17538947.2025.2545583PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN modelZhe Zhou0Hao Wu1Zhenyu Zhang2Bo Kong3Min Yang4Tinghua Ai5Huafei Yu6School of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSchool of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSchool of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSchool of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSchool of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSchool of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSchool of Resource and Environmental Sciences, Wuhan University, Wuhan, People’s Republic of ChinaSaddle-point extraction is essential for accurately identifying topographic features and landforms and conducting geomorphological mapping. However, the widely used positive–negative terrain method (PNTM) is often plagued by a substantial number of false saddle points, a prevalent issue in many extraction techniques. To address this challenge, this study presents a novel model that combines the PNTM with a convolutional neural network (CNN) called PNTM-CNN. In this approach, candidate saddle points are first identified using the PNTM and then refined using a CNN that integrates multiscale topographic features. The experimental results indicate that the PNTM-CNN model, which leverages four scales of features (elevation, aspect, curvature, slope, and hillshade), effectively reduces the occurrence of false saddle points, achieving a precision of 89%, a recall of 83%, and an F1 score of 85%. This performance significantly exceeds that of the traditional moving window analysis and topological association methods. Although the automation level of the PNTM-CNN model requires improvement, the integration of deep learning methods offers new insights for addressing complex topographic feature extraction challenges and shows a promising application potential.https://www.tandfonline.com/doi/10.1080/17538947.2025.2545583Saddle pointspositive–negative terrain methodmultiscale fusionconvolutional neural networks
spellingShingle Zhe Zhou
Hao Wu
Zhenyu Zhang
Bo Kong
Min Yang
Tinghua Ai
Huafei Yu
PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model
International Journal of Digital Earth
Saddle points
positive–negative terrain method
multiscale fusion
convolutional neural networks
title PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model
title_full PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model
title_fullStr PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model
title_full_unstemmed PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model
title_short PNTM-CNN: an approach for saddle-point extraction integrating positive–negative terrain method and multiscale fusion CNN model
title_sort pntm cnn an approach for saddle point extraction integrating positive negative terrain method and multiscale fusion cnn model
topic Saddle points
positive–negative terrain method
multiscale fusion
convolutional neural networks
url https://www.tandfonline.com/doi/10.1080/17538947.2025.2545583
work_keys_str_mv AT zhezhou pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel
AT haowu pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel
AT zhenyuzhang pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel
AT bokong pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel
AT minyang pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel
AT tinghuaai pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel
AT huafeiyu pntmcnnanapproachforsaddlepointextractionintegratingpositivenegativeterrainmethodandmultiscalefusioncnnmodel