Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model
Autonomous underwater garbage grasping and collection pose a great challenge to underwater robots. To assist underwater robots in locating and recognizing underwater garbage objects efficiently, a modified U-Net-based architecture consisting of a deeper contracting path and an expansive path is prop...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/17/6546 |
| _version_ | 1797493089705656320 |
|---|---|
| author | Lifu Wei Shihan Kong Yuquan Wu Junzhi Yu |
| author_facet | Lifu Wei Shihan Kong Yuquan Wu Junzhi Yu |
| author_sort | Lifu Wei |
| collection | DOAJ |
| description | Autonomous underwater garbage grasping and collection pose a great challenge to underwater robots. To assist underwater robots in locating and recognizing underwater garbage objects efficiently, a modified U-Net-based architecture consisting of a deeper contracting path and an expansive path is proposed to accomplish end-to-end image semantic segmentation. In addition, a dataset for underwater garbage semantic segmentation is established. The proposed architecture is further verified in the underwater garbage dataset and the effects of different hyperparameters, loss functions, and optimizers on the performance of refining the predicted segmented mask are examined. It is confirmed that the focal loss function will lead to a boost in solving the target–background unbalance problem. Eventually, the obtained results offer a solid foundation for fast and precise underwater target recognition and operations. |
| first_indexed | 2024-03-10T01:15:00Z |
| format | Article |
| id | doaj.art-420a8a32ab074321946db9eeda2db321 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T01:15:00Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-420a8a32ab074321946db9eeda2db3212023-11-23T14:10:18ZengMDPI AGSensors1424-82202022-08-012217654610.3390/s22176546Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture ModelLifu Wei0Shihan Kong1Yuquan Wu2Junzhi Yu3Department of Advanced Manufacturing and Robotics, College of Engineering, Peking University, Beijing 100871, ChinaDepartment of Advanced Manufacturing and Robotics, College of Engineering, Peking University, Beijing 100871, ChinaInstitute of Software, Chinese Academy of Sciences, Beijing 100190, ChinaDepartment of Advanced Manufacturing and Robotics, College of Engineering, Peking University, Beijing 100871, ChinaAutonomous underwater garbage grasping and collection pose a great challenge to underwater robots. To assist underwater robots in locating and recognizing underwater garbage objects efficiently, a modified U-Net-based architecture consisting of a deeper contracting path and an expansive path is proposed to accomplish end-to-end image semantic segmentation. In addition, a dataset for underwater garbage semantic segmentation is established. The proposed architecture is further verified in the underwater garbage dataset and the effects of different hyperparameters, loss functions, and optimizers on the performance of refining the predicted segmented mask are examined. It is confirmed that the focal loss function will lead to a boost in solving the target–background unbalance problem. Eventually, the obtained results offer a solid foundation for fast and precise underwater target recognition and operations.https://www.mdpi.com/1424-8220/22/17/6546underwater garbage collectionsemantic segmentationdeep learningU-Net |
| spellingShingle | Lifu Wei Shihan Kong Yuquan Wu Junzhi Yu Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model Sensors underwater garbage collection semantic segmentation deep learning U-Net |
| title | Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model |
| title_full | Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model |
| title_fullStr | Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model |
| title_full_unstemmed | Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model |
| title_short | Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model |
| title_sort | image semantic segmentation of underwater garbage with modified u net architecture model |
| topic | underwater garbage collection semantic segmentation deep learning U-Net |
| url | https://www.mdpi.com/1424-8220/22/17/6546 |
| work_keys_str_mv | AT lifuwei imagesemanticsegmentationofunderwatergarbagewithmodifiedunetarchitecturemodel AT shihankong imagesemanticsegmentationofunderwatergarbagewithmodifiedunetarchitecturemodel AT yuquanwu imagesemanticsegmentationofunderwatergarbagewithmodifiedunetarchitecturemodel AT junzhiyu imagesemanticsegmentationofunderwatergarbagewithmodifiedunetarchitecturemodel |