Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images
Riverbank sand overexploitation is threatening the ecology and shipping safety of rivers. The rapid identification of riverbank sand mining areas from satellite images is extremely important for ecological protection and shipping management. Image segmentation methods based on AI technology are grad...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/2/227 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587599889301504 |
|---|---|
| author | Hongyang Zhang Shuo Liu Huamei Liu |
| author_facet | Hongyang Zhang Shuo Liu Huamei Liu |
| author_sort | Hongyang Zhang |
| collection | DOAJ |
| description | Riverbank sand overexploitation is threatening the ecology and shipping safety of rivers. The rapid identification of riverbank sand mining areas from satellite images is extremely important for ecological protection and shipping management. Image segmentation methods based on AI technology are gradually becoming popular in academia and industry. However, traditional neural networks have complex structures and numerous parameters, making them unsuitable for meeting the needs of rapid extraction in large areas. To improve efficiency, we proposed a lightweight multi-scale network (LMS Net), which uses a lightweight multi-scale (LMS) block in both the encoder and decoder. The lightweight multi-scale block combines parallel computing and depthwise convolution to reduce the parameters of the network and enhance its multi-scale extraction ability. We created a benchmark dataset to validate the accuracy and efficiency improvements of our network. Comparative experiments and ablation studies proved that our LMS Net is more efficient than traditional methods like Unet and more accurate than typical lightweight methods like Ghostnet and other more recent methods. The performance of our proposed network meets the requirements of river management. |
| format | Article |
| id | doaj-art-d47a80e1b58b4ccbb001dfd82494953b |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-d47a80e1b58b4ccbb001dfd82494953b2025-01-24T13:47:47ZengMDPI AGRemote Sensing2072-42922025-01-0117222710.3390/rs17020227Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite ImagesHongyang Zhang0Shuo Liu1Huamei Liu2Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaRiverbank sand overexploitation is threatening the ecology and shipping safety of rivers. The rapid identification of riverbank sand mining areas from satellite images is extremely important for ecological protection and shipping management. Image segmentation methods based on AI technology are gradually becoming popular in academia and industry. However, traditional neural networks have complex structures and numerous parameters, making them unsuitable for meeting the needs of rapid extraction in large areas. To improve efficiency, we proposed a lightweight multi-scale network (LMS Net), which uses a lightweight multi-scale (LMS) block in both the encoder and decoder. The lightweight multi-scale block combines parallel computing and depthwise convolution to reduce the parameters of the network and enhance its multi-scale extraction ability. We created a benchmark dataset to validate the accuracy and efficiency improvements of our network. Comparative experiments and ablation studies proved that our LMS Net is more efficient than traditional methods like Unet and more accurate than typical lightweight methods like Ghostnet and other more recent methods. The performance of our proposed network meets the requirements of river management.https://www.mdpi.com/2072-4292/17/2/227riverbanksand miningsegmentationlightweightdeep learningneural network |
| spellingShingle | Hongyang Zhang Shuo Liu Huamei Liu Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images Remote Sensing riverbank sand mining segmentation lightweight deep learning neural network |
| title | Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images |
| title_full | Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images |
| title_fullStr | Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images |
| title_full_unstemmed | Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images |
| title_short | Lightweight Multi-Scale Network for Segmentation of Riverbank Sand Mining Area in Satellite Images |
| title_sort | lightweight multi scale network for segmentation of riverbank sand mining area in satellite images |
| topic | riverbank sand mining segmentation lightweight deep learning neural network |
| url | https://www.mdpi.com/2072-4292/17/2/227 |
| work_keys_str_mv | AT hongyangzhang lightweightmultiscalenetworkforsegmentationofriverbanksandminingareainsatelliteimages AT shuoliu lightweightmultiscalenetworkforsegmentationofriverbanksandminingareainsatelliteimages AT huameiliu lightweightmultiscalenetworkforsegmentationofriverbanksandminingareainsatelliteimages |