An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection

This study presents AD-MRCNN, an advanced deep learning framework for automated atmospheric front detection that addresses two critical limitations in existing methods. First, current approaches directly input raw meteorological data without optimizing feature compatibility, potentially hindering mo...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinya Ding, Xuan Peng, Yanguang Xue, Liang Zhang, Tianying Wang, Yunpeng Zhang
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/14/7855
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849714477785677824
author Xinya Ding
Xuan Peng
Yanguang Xue
Liang Zhang
Tianying Wang
Yunpeng Zhang
author_facet Xinya Ding
Xuan Peng
Yanguang Xue
Liang Zhang
Tianying Wang
Yunpeng Zhang
author_sort Xinya Ding
collection DOAJ
description This study presents AD-MRCNN, an advanced deep learning framework for automated atmospheric front detection that addresses two critical limitations in existing methods. First, current approaches directly input raw meteorological data without optimizing feature compatibility, potentially hindering model performance. Second, they typically only provide frontal category information without identifying individual frontal systems. Our solution integrates two key innovations: 1. An intelligent adapter module that performs adaptive feature fusion, automatically weighting and combining multi-source meteorological inputs (including temperature, wind fields, and humidity data) to maximize their synergistic effects while minimizing feature conflicts; the utilized network achieves an average improvement of over 4% across various metrics. 2. An enhanced instance segmentation network based on Mask R-CNN architecture that simultaneously achieves (1) precise frontal type classification (cold/warm/stationary/occluded), (2) accurate spatial localization, and (3) identification of distinct frontal systems. Comprehensive evaluation using ERA5 reanalysis data (2009–2018) demonstrates significant improvements, including an 85.1% F1-score, outperforming traditional methods (TFP: 63.1%) and deep learning approaches (Unet: 83.3%), and a 31% reduction in false alarms compared to semantic segmentation methods. The framework’s modular design allows for potential application to other meteorological feature detection tasks. Future work will focus on incorporating temporal dynamics for frontal evolution prediction.
format Article
id doaj-art-21a256eb864b488caae950fe9aaab933
institution DOAJ
issn 2076-3417
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-21a256eb864b488caae950fe9aaab9332025-08-20T03:13:42ZengMDPI AGApplied Sciences2076-34172025-07-011514785510.3390/app15147855An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front DetectionXinya Ding0Xuan Peng1Yanguang Xue2Liang Zhang3Tianying Wang4Yunpeng Zhang5Key Laboratory of Smart Earth, Beijing 100029, ChinaKey Laboratory of Smart Earth, Beijing 100029, ChinaKey Laboratory of Smart Earth, Beijing 100029, ChinaThe College of Meteorology and Oceanography, National University of Defense Technology, Changsha 410005, ChinaInstitute of Meteorological Sciences of Hunan Province, Changsha 410118, ChinaThe College of Meteorology and Oceanography, National University of Defense Technology, Changsha 410005, ChinaThis study presents AD-MRCNN, an advanced deep learning framework for automated atmospheric front detection that addresses two critical limitations in existing methods. First, current approaches directly input raw meteorological data without optimizing feature compatibility, potentially hindering model performance. Second, they typically only provide frontal category information without identifying individual frontal systems. Our solution integrates two key innovations: 1. An intelligent adapter module that performs adaptive feature fusion, automatically weighting and combining multi-source meteorological inputs (including temperature, wind fields, and humidity data) to maximize their synergistic effects while minimizing feature conflicts; the utilized network achieves an average improvement of over 4% across various metrics. 2. An enhanced instance segmentation network based on Mask R-CNN architecture that simultaneously achieves (1) precise frontal type classification (cold/warm/stationary/occluded), (2) accurate spatial localization, and (3) identification of distinct frontal systems. Comprehensive evaluation using ERA5 reanalysis data (2009–2018) demonstrates significant improvements, including an 85.1% F1-score, outperforming traditional methods (TFP: 63.1%) and deep learning approaches (Unet: 83.3%), and a 31% reduction in false alarms compared to semantic segmentation methods. The framework’s modular design allows for potential application to other meteorological feature detection tasks. Future work will focus on incorporating temporal dynamics for frontal evolution prediction.https://www.mdpi.com/2076-3417/15/14/7855atmospheric frontautomatic identificationadaptersegmentation networkAD-MRCNN model
spellingShingle Xinya Ding
Xuan Peng
Yanguang Xue
Liang Zhang
Tianying Wang
Yunpeng Zhang
An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection
Applied Sciences
atmospheric front
automatic identification
adapter
segmentation network
AD-MRCNN model
title An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection
title_full An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection
title_fullStr An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection
title_full_unstemmed An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection
title_short An Adapter and Segmentation Network-Based Approach for Automated Atmospheric Front Detection
title_sort adapter and segmentation network based approach for automated atmospheric front detection
topic atmospheric front
automatic identification
adapter
segmentation network
AD-MRCNN model
url https://www.mdpi.com/2076-3417/15/14/7855
work_keys_str_mv AT xinyading anadapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT xuanpeng anadapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT yanguangxue anadapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT liangzhang anadapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT tianyingwang anadapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT yunpengzhang anadapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT xinyading adapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT xuanpeng adapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT yanguangxue adapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT liangzhang adapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT tianyingwang adapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection
AT yunpengzhang adapterandsegmentationnetworkbasedapproachforautomatedatmosphericfrontdetection