Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery
The traditional paradigm for studying urban morphology involves the interpretation of Nolli maps, using methods such as morphometrics and visual neural networks. Previous studies on urban morphology discovery have always been based on raster analysis and have been limited to the central city area. R...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/14/5/196 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849327321484361728 |
|---|---|
| author | Chunliang Hua Daijun Chen Mengyuan Niu Lizhong Gao Junyan Yang Qiao Wang |
| author_facet | Chunliang Hua Daijun Chen Mengyuan Niu Lizhong Gao Junyan Yang Qiao Wang |
| author_sort | Chunliang Hua |
| collection | DOAJ |
| description | The traditional paradigm for studying urban morphology involves the interpretation of Nolli maps, using methods such as morphometrics and visual neural networks. Previous studies on urban morphology discovery have always been based on raster analysis and have been limited to the central city area. Raster analysis can lead to fragmented forms, and focusing only on the central city area ignores many representative urban forms in the suburbs and towns. In this study, a vast and complex dataset was applied to the urban morphology discovery based on the administrative community or village boundary, and a new image deformation pipeline was proposed to enhance the morphological characteristics of building groups. This allows visual neural networks to focus on extracting the morphological characteristics of building groups. Additionally, the research on urban morphology often uses unsupervised learning, which means that the learning process is difficult to control. Therefore, we refined the NT-Xent loss so that it can integrate morphological indicators. This improvement allows the visual neural network to “recognize” the similarity of samples during optimization. By defining the similarity, we can guide the network to bring samples closer or move them farther apart based on certain morphological indicators. Three Chinese cities were used for our testing. Representative urban types were identified, particularly some types located at the urban fringe. The data analysis demonstrated the effectiveness of our image deformation pipeline and loss function, and the sociological analysis illustrated the unique urban functions of these urban types. |
| format | Article |
| id | doaj-art-5047ebdd83f048ad80bd9c1a395cf30c |
| institution | Kabale University |
| issn | 2220-9964 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj-art-5047ebdd83f048ad80bd9c1a395cf30c2025-08-20T03:47:54ZengMDPI AGISPRS International Journal of Geo-Information2220-99642025-05-0114519610.3390/ijgi14050196Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology DiscoveryChunliang Hua0Daijun Chen1Mengyuan Niu2Lizhong Gao3Junyan Yang4Qiao Wang5School of Information Science and Engineering, Southeast University, Nanjing 211189, ChinaSchool of Architecture, Southeast University, Nanjing 210096, ChinaSchool of Information Science and Engineering, Southeast University, Nanjing 211189, ChinaSchool of Information Science and Engineering, Southeast University, Nanjing 211189, ChinaSchool of Architecture, Southeast University, Nanjing 210096, ChinaSchool of Information Science and Engineering, Southeast University, Nanjing 211189, ChinaThe traditional paradigm for studying urban morphology involves the interpretation of Nolli maps, using methods such as morphometrics and visual neural networks. Previous studies on urban morphology discovery have always been based on raster analysis and have been limited to the central city area. Raster analysis can lead to fragmented forms, and focusing only on the central city area ignores many representative urban forms in the suburbs and towns. In this study, a vast and complex dataset was applied to the urban morphology discovery based on the administrative community or village boundary, and a new image deformation pipeline was proposed to enhance the morphological characteristics of building groups. This allows visual neural networks to focus on extracting the morphological characteristics of building groups. Additionally, the research on urban morphology often uses unsupervised learning, which means that the learning process is difficult to control. Therefore, we refined the NT-Xent loss so that it can integrate morphological indicators. This improvement allows the visual neural network to “recognize” the similarity of samples during optimization. By defining the similarity, we can guide the network to bring samples closer or move them farther apart based on certain morphological indicators. Three Chinese cities were used for our testing. Representative urban types were identified, particularly some types located at the urban fringe. The data analysis demonstrated the effectiveness of our image deformation pipeline and loss function, and the sociological analysis illustrated the unique urban functions of these urban types.https://www.mdpi.com/2220-9964/14/5/196urban formNolli mapNT-Xent lossunsupervised learningimage deformationcontrastive learning |
| spellingShingle | Chunliang Hua Daijun Chen Mengyuan Niu Lizhong Gao Junyan Yang Qiao Wang Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery ISPRS International Journal of Geo-Information urban form Nolli map NT-Xent loss unsupervised learning image deformation contrastive learning |
| title | Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery |
| title_full | Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery |
| title_fullStr | Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery |
| title_full_unstemmed | Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery |
| title_short | Contrastive Learning with Image Deformation and Refined NT-Xent Loss for Urban Morphology Discovery |
| title_sort | contrastive learning with image deformation and refined nt xent loss for urban morphology discovery |
| topic | urban form Nolli map NT-Xent loss unsupervised learning image deformation contrastive learning |
| url | https://www.mdpi.com/2220-9964/14/5/196 |
| work_keys_str_mv | AT chunlianghua contrastivelearningwithimagedeformationandrefinedntxentlossforurbanmorphologydiscovery AT daijunchen contrastivelearningwithimagedeformationandrefinedntxentlossforurbanmorphologydiscovery AT mengyuanniu contrastivelearningwithimagedeformationandrefinedntxentlossforurbanmorphologydiscovery AT lizhonggao contrastivelearningwithimagedeformationandrefinedntxentlossforurbanmorphologydiscovery AT junyanyang contrastivelearningwithimagedeformationandrefinedntxentlossforurbanmorphologydiscovery AT qiaowang contrastivelearningwithimagedeformationandrefinedntxentlossforurbanmorphologydiscovery |