Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions

Urban green spaces are an important part of the urban ecosystem and hold significant ecological value. To effectively protect these green spaces, urban managers urgently need to identify them and monitor their changes. Common urban vegetation positioning methods use deep learning segmentation models...

Full description

Saved in:
Bibliographic Details
Main Authors: Hanzhang Liu, Shijie Yang, Chengwu Long, Jiateng Yuan, Qirui Yang, Jiahua Fan, Bingnan Meng, Zhibo Chen, Fu Xu, Chao Mou
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/12/2058
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849704744591818752
author Hanzhang Liu
Shijie Yang
Chengwu Long
Jiateng Yuan
Qirui Yang
Jiahua Fan
Bingnan Meng
Zhibo Chen
Fu Xu
Chao Mou
author_facet Hanzhang Liu
Shijie Yang
Chengwu Long
Jiateng Yuan
Qirui Yang
Jiahua Fan
Bingnan Meng
Zhibo Chen
Fu Xu
Chao Mou
author_sort Hanzhang Liu
collection DOAJ
description Urban green spaces are an important part of the urban ecosystem and hold significant ecological value. To effectively protect these green spaces, urban managers urgently need to identify them and monitor their changes. Common urban vegetation positioning methods use deep learning segmentation models to process street view data in urban areas, but this is usually inefficient and inaccurate. The main reason is that they are not applicable to the variable climate of urban scenarios, especially performing poorly in adverse weather conditions such as heavy fog that are common in cities. Additionally, these algorithms also have performance limitations such as inaccurate boundary area positioning. To address these challenges, we propose the UGSAM method that utilizes the high-performance multimodal large language model, the Segment Anything Model (i.e., SAM). In the UGSAM, a dual-branch defogging network WRPM is incorporated, which consists of the dense fog network FFA-Net, the light fog network LS-UNet, and the feature fusion network FIM, achieving precise identification of vegetation areas in adverse urban weather conditions. Moreover, we have designed a micro-correction network SCP-Net suitable for specific urban scenarios to further improve the accuracy of urban vegetation positioning. The UGSAM was compared with three classic deep learning algorithms and the SAM. Experimental results show that under adverse weather conditions, the UGSAM performs best in OA (0.8615), mIoU (0.8490), recall (0.9345), and precision (0.9027), surpassing the baseline model FCN (OA improvement 28.19%) and PointNet++ (OA improvement 30.02%). Compared with the SAM, the UGSAM improves the segmentation accuracy by 16.29% under adverse weather conditions and by 1.03% under good weather conditions. This method is expected to play a key role in the analysis of urban green spaces under adverse weather conditions and provide innovative insights for urban development.
format Article
id doaj-art-72d778b7a47b487f970f9c1b46f1cdef
institution DOAJ
issn 2072-4292
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-72d778b7a47b487f970f9c1b46f1cdef2025-08-20T03:16:39ZengMDPI AGRemote Sensing2072-42922025-06-011712205810.3390/rs17122058Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather ConditionsHanzhang Liu0Shijie Yang1Chengwu Long2Jiateng Yuan3Qirui Yang4Jiahua Fan5Bingnan Meng6Zhibo Chen7Fu Xu8Chao Mou9School of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaSchool of Computer Science and Technology, East China Normal University, Shanghai 200062, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaGeneral Forestry Station of Beijing Municipality, Beijing 100029, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaSchool of Information Science and Technology (School of Artificial Intelligence), Beijing Forestry University, Beijing 100083, ChinaUrban green spaces are an important part of the urban ecosystem and hold significant ecological value. To effectively protect these green spaces, urban managers urgently need to identify them and monitor their changes. Common urban vegetation positioning methods use deep learning segmentation models to process street view data in urban areas, but this is usually inefficient and inaccurate. The main reason is that they are not applicable to the variable climate of urban scenarios, especially performing poorly in adverse weather conditions such as heavy fog that are common in cities. Additionally, these algorithms also have performance limitations such as inaccurate boundary area positioning. To address these challenges, we propose the UGSAM method that utilizes the high-performance multimodal large language model, the Segment Anything Model (i.e., SAM). In the UGSAM, a dual-branch defogging network WRPM is incorporated, which consists of the dense fog network FFA-Net, the light fog network LS-UNet, and the feature fusion network FIM, achieving precise identification of vegetation areas in adverse urban weather conditions. Moreover, we have designed a micro-correction network SCP-Net suitable for specific urban scenarios to further improve the accuracy of urban vegetation positioning. The UGSAM was compared with three classic deep learning algorithms and the SAM. Experimental results show that under adverse weather conditions, the UGSAM performs best in OA (0.8615), mIoU (0.8490), recall (0.9345), and precision (0.9027), surpassing the baseline model FCN (OA improvement 28.19%) and PointNet++ (OA improvement 30.02%). Compared with the SAM, the UGSAM improves the segmentation accuracy by 16.29% under adverse weather conditions and by 1.03% under good weather conditions. This method is expected to play a key role in the analysis of urban green spaces under adverse weather conditions and provide innovative insights for urban development.https://www.mdpi.com/2072-4292/17/12/2058urban green spacesstreetscape imagerymultimodal large language modeldeep learningSegment Anything Model
spellingShingle Hanzhang Liu
Shijie Yang
Chengwu Long
Jiateng Yuan
Qirui Yang
Jiahua Fan
Bingnan Meng
Zhibo Chen
Fu Xu
Chao Mou
Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions
Remote Sensing
urban green spaces
streetscape imagery
multimodal large language model
deep learning
Segment Anything Model
title Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions
title_full Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions
title_fullStr Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions
title_full_unstemmed Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions
title_short Urban Greening Analysis: A Multimodal Large Language Model for Pinpointing Vegetation Areas in Adverse Weather Conditions
title_sort urban greening analysis a multimodal large language model for pinpointing vegetation areas in adverse weather conditions
topic urban green spaces
streetscape imagery
multimodal large language model
deep learning
Segment Anything Model
url https://www.mdpi.com/2072-4292/17/12/2058
work_keys_str_mv AT hanzhangliu urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT shijieyang urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT chengwulong urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT jiatengyuan urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT qiruiyang urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT jiahuafan urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT bingnanmeng urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT zhibochen urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT fuxu urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions
AT chaomou urbangreeninganalysisamultimodallargelanguagemodelforpinpointingvegetationareasinadverseweatherconditions