Fine-Grained Building Classification in Rural Areas Based on GF-7 Data
Building type information is widely used in various fields, such as disaster management, urbanization studies, and population modelling. Few studies have been conducted on fine-grained building classification in rural areas using China’s Gaofen-7 (GF-7) high-resolution stereo mapping satellite data....
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/2/392 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587497419309056 |
|---|---|
| author | Mingbo Liu Ping Wang Peng Han Longfei Liu Baotian Li |
| author_facet | Mingbo Liu Ping Wang Peng Han Longfei Liu Baotian Li |
| author_sort | Mingbo Liu |
| collection | DOAJ |
| description | Building type information is widely used in various fields, such as disaster management, urbanization studies, and population modelling. Few studies have been conducted on fine-grained building classification in rural areas using China’s Gaofen-7 (GF-7) high-resolution stereo mapping satellite data. In this study, we employed a two-stage method combining supervised classification and unsupervised clustering to classify buildings in the rural area of Pingquan, northern China, based on building footprints, building heights, and multispectral information extracted from GF-7 data. In the supervised classification stage, we compared different classification models, including Extreme Gradient Boosting (XGBoost) and Random Forest classifiers. The best-performing XGBoost model achieved an overall roof type classification accuracy of 88.89%. Additionally, we proposed a template-based building height correction method for pitched roof buildings, which combined geometric features of the building footprint, street view photos, and height information extracted from the GF-7 stereo image. This method reduced the RMSE of the pitched roof building heights from 2.28 m to 1.20 m. In the cluster analysis stage, buildings with different roof types were further classified in the color and shape feature spaces and combined with the building height information to produce fine-grained building type codes. The results of the roof type classification and fine-grained building classification reveal the physical and geometric characteristics of buildings and the spatial distribution of different building types in the study area. The building classification method proposed in this study has broad application prospects for disaster management in rural areas. |
| format | Article |
| id | doaj-art-5d71a57d25c745f6a04f7e61f026afb8 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-5d71a57d25c745f6a04f7e61f026afb82025-01-24T13:48:45ZengMDPI AGSensors1424-82202025-01-0125239210.3390/s25020392Fine-Grained Building Classification in Rural Areas Based on GF-7 DataMingbo Liu0Ping Wang1Peng Han2Longfei Liu3Baotian Li4National Disaster Reduction Center of China, Ministry of Emergency Management of the People’s Republic of China, Beijing 100124, ChinaNational Disaster Reduction Center of China, Ministry of Emergency Management of the People’s Republic of China, Beijing 100124, ChinaNational Disaster Reduction Center of China, Ministry of Emergency Management of the People’s Republic of China, Beijing 100124, ChinaNational Disaster Reduction Center of China, Ministry of Emergency Management of the People’s Republic of China, Beijing 100124, ChinaBureau of Emergency Management of Pingquan City, Pingquan 067500, ChinaBuilding type information is widely used in various fields, such as disaster management, urbanization studies, and population modelling. Few studies have been conducted on fine-grained building classification in rural areas using China’s Gaofen-7 (GF-7) high-resolution stereo mapping satellite data. In this study, we employed a two-stage method combining supervised classification and unsupervised clustering to classify buildings in the rural area of Pingquan, northern China, based on building footprints, building heights, and multispectral information extracted from GF-7 data. In the supervised classification stage, we compared different classification models, including Extreme Gradient Boosting (XGBoost) and Random Forest classifiers. The best-performing XGBoost model achieved an overall roof type classification accuracy of 88.89%. Additionally, we proposed a template-based building height correction method for pitched roof buildings, which combined geometric features of the building footprint, street view photos, and height information extracted from the GF-7 stereo image. This method reduced the RMSE of the pitched roof building heights from 2.28 m to 1.20 m. In the cluster analysis stage, buildings with different roof types were further classified in the color and shape feature spaces and combined with the building height information to produce fine-grained building type codes. The results of the roof type classification and fine-grained building classification reveal the physical and geometric characteristics of buildings and the spatial distribution of different building types in the study area. The building classification method proposed in this study has broad application prospects for disaster management in rural areas.https://www.mdpi.com/1424-8220/25/2/392GF-7 databuilding classificationbuilding heightrural areas |
| spellingShingle | Mingbo Liu Ping Wang Peng Han Longfei Liu Baotian Li Fine-Grained Building Classification in Rural Areas Based on GF-7 Data Sensors GF-7 data building classification building height rural areas |
| title | Fine-Grained Building Classification in Rural Areas Based on GF-7 Data |
| title_full | Fine-Grained Building Classification in Rural Areas Based on GF-7 Data |
| title_fullStr | Fine-Grained Building Classification in Rural Areas Based on GF-7 Data |
| title_full_unstemmed | Fine-Grained Building Classification in Rural Areas Based on GF-7 Data |
| title_short | Fine-Grained Building Classification in Rural Areas Based on GF-7 Data |
| title_sort | fine grained building classification in rural areas based on gf 7 data |
| topic | GF-7 data building classification building height rural areas |
| url | https://www.mdpi.com/1424-8220/25/2/392 |
| work_keys_str_mv | AT mingboliu finegrainedbuildingclassificationinruralareasbasedongf7data AT pingwang finegrainedbuildingclassificationinruralareasbasedongf7data AT penghan finegrainedbuildingclassificationinruralareasbasedongf7data AT longfeiliu finegrainedbuildingclassificationinruralareasbasedongf7data AT baotianli finegrainedbuildingclassificationinruralareasbasedongf7data |