Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance
Building change detection has always been an important research focus in production and urbanization. In recent years, deep learning methods have demonstrated a powerful ability in the field of detecting remote sensing changes. However, due to the heterogeneity of remote sensing and the characterist...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/20/4171 |
| _version_ | 1797513181765042176 |
|---|---|
| author | Junkang Xue Hao Xu Hui Yang Biao Wang Penghai Wu Jaewan Choi Lixiao Cai Yanlan Wu |
| author_facet | Junkang Xue Hao Xu Hui Yang Biao Wang Penghai Wu Jaewan Choi Lixiao Cai Yanlan Wu |
| author_sort | Junkang Xue |
| collection | DOAJ |
| description | Building change detection has always been an important research focus in production and urbanization. In recent years, deep learning methods have demonstrated a powerful ability in the field of detecting remote sensing changes. However, due to the heterogeneity of remote sensing and the characteristics of buildings, the current methods do not present an effective means to perceive building changes or the ability to fuse multi-temporal remote sensing features, which leads to fragmented and incomplete results. In this article, we propose a multi-branched network structure to fuse the semantic information of the building changes at different levels. In this model, two accessory branches were used to guide the buildings’ semantic information under different time sequences, and the main branches can merge the change information. In addition, we also designed a feature enhancement layer to further strengthen the integration of the main and accessory branch information. For ablation experiments, we designed experiments on the above optimization process. For MDEFNET, we designed experiments which compare with typical deep learning model and recent deep learning change detection methods. Experimentation with the WHU Building Change Detection Dataset showed that the method in this paper obtained accuracies of 0.8526, 0.9418, and 0.9204 in Intersection over Union (IoU), Recall, and F1 Score, respectively, which could assess building change areas with complete boundaries and accurate results. |
| first_indexed | 2024-03-10T06:13:03Z |
| format | Article |
| id | doaj.art-273def44022f414da3411a5d5b6660ce |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T06:13:03Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-273def44022f414da3411a5d5b6660ce2023-11-22T19:55:16ZengMDPI AGRemote Sensing2072-42922021-10-011320417110.3390/rs13204171Multi-Feature Enhanced Building Change Detection Based on Semantic Information GuidanceJunkang Xue0Hao Xu1Hui Yang2Biao Wang3Penghai Wu4Jaewan Choi5Lixiao Cai6Yanlan Wu7School of Resources and Environmental Engineering, Anhui University, Hefei 230601, ChinaInstitute of Spacecraft System Engineering, Beijing 100094, ChinaSchool of Resources and Environmental Engineering, Anhui University, Hefei 230601, ChinaSchool of Resources and Environmental Engineering, Anhui University, Hefei 230601, ChinaSchool of Resources and Environmental Engineering, Anhui University, Hefei 230601, ChinaSchool of Civil Engineering, Chungbuk National University, Chungju 28644, KoreaSchool of Design Group, Shandong Jianzhu University, Jinan 250101, ChinaSchool of Resources and Environmental Engineering, Anhui University, Hefei 230601, ChinaBuilding change detection has always been an important research focus in production and urbanization. In recent years, deep learning methods have demonstrated a powerful ability in the field of detecting remote sensing changes. However, due to the heterogeneity of remote sensing and the characteristics of buildings, the current methods do not present an effective means to perceive building changes or the ability to fuse multi-temporal remote sensing features, which leads to fragmented and incomplete results. In this article, we propose a multi-branched network structure to fuse the semantic information of the building changes at different levels. In this model, two accessory branches were used to guide the buildings’ semantic information under different time sequences, and the main branches can merge the change information. In addition, we also designed a feature enhancement layer to further strengthen the integration of the main and accessory branch information. For ablation experiments, we designed experiments on the above optimization process. For MDEFNET, we designed experiments which compare with typical deep learning model and recent deep learning change detection methods. Experimentation with the WHU Building Change Detection Dataset showed that the method in this paper obtained accuracies of 0.8526, 0.9418, and 0.9204 in Intersection over Union (IoU), Recall, and F1 Score, respectively, which could assess building change areas with complete boundaries and accurate results.https://www.mdpi.com/2072-4292/13/20/4171remote sensingbuilding change detectiondeep learningmulti-branchsemantic information |
| spellingShingle | Junkang Xue Hao Xu Hui Yang Biao Wang Penghai Wu Jaewan Choi Lixiao Cai Yanlan Wu Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance Remote Sensing remote sensing building change detection deep learning multi-branch semantic information |
| title | Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance |
| title_full | Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance |
| title_fullStr | Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance |
| title_full_unstemmed | Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance |
| title_short | Multi-Feature Enhanced Building Change Detection Based on Semantic Information Guidance |
| title_sort | multi feature enhanced building change detection based on semantic information guidance |
| topic | remote sensing building change detection deep learning multi-branch semantic information |
| url | https://www.mdpi.com/2072-4292/13/20/4171 |
| work_keys_str_mv | AT junkangxue multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT haoxu multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT huiyang multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT biaowang multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT penghaiwu multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT jaewanchoi multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT lixiaocai multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance AT yanlanwu multifeatureenhancedbuildingchangedetectionbasedonsemanticinformationguidance |