Deep Fusion of DOM and DSM Features for Benggang Discovery
Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cann...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-08-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/10/8/556 |
| _version_ | 1797523675026554880 |
|---|---|
| author | Shengyu Shen Jiasheng Chen Shaoyi Zhang Dongbing Cheng Zhigang Wang Tong Zhang |
| author_facet | Shengyu Shen Jiasheng Chen Shaoyi Zhang Dongbing Cheng Zhigang Wang Tong Zhang |
| author_sort | Shengyu Shen |
| collection | DOAJ |
| description | Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks. |
| first_indexed | 2024-03-10T08:46:23Z |
| format | Article |
| id | doaj.art-f6cafb7cb7374a818f0afc026a7bdda9 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-10T08:46:23Z |
| publishDate | 2021-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-f6cafb7cb7374a818f0afc026a7bdda92023-11-22T07:53:38ZengMDPI AGISPRS International Journal of Geo-Information2220-99642021-08-0110855610.3390/ijgi10080556Deep Fusion of DOM and DSM Features for Benggang DiscoveryShengyu Shen0Jiasheng Chen1Shaoyi Zhang2Dongbing Cheng3Zhigang Wang4Tong Zhang5Department of Soil and Water Conservation, Changjiang River Scientific Research Institute (CRSRI), Wuhan 430010, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan University, Wuhan 430079, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaDepartment of Soil and Water Conservation, Changjiang River Scientific Research Institute (CRSRI), Wuhan 430010, ChinaDepartment of Soil and Water Conservation, Changjiang River Scientific Research Institute (CRSRI), Wuhan 430010, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan University, Wuhan 430079, ChinaBenggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks.https://www.mdpi.com/2220-9964/10/8/556benggangdeep learningfusionCNNDOMDSM |
| spellingShingle | Shengyu Shen Jiasheng Chen Shaoyi Zhang Dongbing Cheng Zhigang Wang Tong Zhang Deep Fusion of DOM and DSM Features for Benggang Discovery ISPRS International Journal of Geo-Information benggang deep learning fusion CNN DOM DSM |
| title | Deep Fusion of DOM and DSM Features for Benggang Discovery |
| title_full | Deep Fusion of DOM and DSM Features for Benggang Discovery |
| title_fullStr | Deep Fusion of DOM and DSM Features for Benggang Discovery |
| title_full_unstemmed | Deep Fusion of DOM and DSM Features for Benggang Discovery |
| title_short | Deep Fusion of DOM and DSM Features for Benggang Discovery |
| title_sort | deep fusion of dom and dsm features for benggang discovery |
| topic | benggang deep learning fusion CNN DOM DSM |
| url | https://www.mdpi.com/2220-9964/10/8/556 |
| work_keys_str_mv | AT shengyushen deepfusionofdomanddsmfeaturesforbenggangdiscovery AT jiashengchen deepfusionofdomanddsmfeaturesforbenggangdiscovery AT shaoyizhang deepfusionofdomanddsmfeaturesforbenggangdiscovery AT dongbingcheng deepfusionofdomanddsmfeaturesforbenggangdiscovery AT zhigangwang deepfusionofdomanddsmfeaturesforbenggangdiscovery AT tongzhang deepfusionofdomanddsmfeaturesforbenggangdiscovery |