ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance
Obstacle detection for autonomous navigation through semantic image segmentation using neural networks has grown in popularity for use in unmanned ground and surface vehicles because of its ability to rapidly create a highly accurate pixel-wise classification of complex scenes. Due to the lack of av...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/13/4681 |
| _version_ | 1827598590965448704 |
|---|---|
| author | Reeve Lambert Jalil Chavez-Galaviz Jianwen Li Nina Mahmoudian |
| author_facet | Reeve Lambert Jalil Chavez-Galaviz Jianwen Li Nina Mahmoudian |
| author_sort | Reeve Lambert |
| collection | DOAJ |
| description | Obstacle detection for autonomous navigation through semantic image segmentation using neural networks has grown in popularity for use in unmanned ground and surface vehicles because of its ability to rapidly create a highly accurate pixel-wise classification of complex scenes. Due to the lack of available training data, semantic networks are rarely applied to navigation in complex water scenes such as rivers, creeks, canals, and harbors. This work seeks to address the issue by making a one-of-its-kind River Obstacle Segmentation En-Route By USV Dataset (ROSEBUD) publicly available for use in robotic SLAM applications that map water and non-water entities in fluvial images from the water level. ROSEBUD provides a challenging baseline for surface navigation in complex environments using complex fluvial scenes. The dataset contains 549 images encompassing various water qualities, seasons, and obstacle types that were taken on narrow inland rivers and then hand annotated for use in semantic network training. The difference between the ROSEBUD dataset and existing marine datasets was verified. Two state-of-the-art networks were trained on existing water segmentation datasets and tested for generalization to the ROSEBUD dataset. Results from further training show that modern semantic networks custom made for water recognition, and trained on marine images, can properly segment large areas, but they struggle to properly segment small obstacles in fluvial scenes without further training on the ROSEBUD dataset. |
| first_indexed | 2024-03-09T03:55:12Z |
| format | Article |
| id | doaj.art-0e3ceee294b24390a15aaabc75ae6dc1 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T03:55:12Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0e3ceee294b24390a15aaabc75ae6dc12023-12-03T14:21:35ZengMDPI AGSensors1424-82202022-06-012213468110.3390/s22134681ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle AvoidanceReeve Lambert0Jalil Chavez-Galaviz1Jianwen Li2Nina Mahmoudian3The School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USAThe School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USAThe School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USAThe School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USAObstacle detection for autonomous navigation through semantic image segmentation using neural networks has grown in popularity for use in unmanned ground and surface vehicles because of its ability to rapidly create a highly accurate pixel-wise classification of complex scenes. Due to the lack of available training data, semantic networks are rarely applied to navigation in complex water scenes such as rivers, creeks, canals, and harbors. This work seeks to address the issue by making a one-of-its-kind River Obstacle Segmentation En-Route By USV Dataset (ROSEBUD) publicly available for use in robotic SLAM applications that map water and non-water entities in fluvial images from the water level. ROSEBUD provides a challenging baseline for surface navigation in complex environments using complex fluvial scenes. The dataset contains 549 images encompassing various water qualities, seasons, and obstacle types that were taken on narrow inland rivers and then hand annotated for use in semantic network training. The difference between the ROSEBUD dataset and existing marine datasets was verified. Two state-of-the-art networks were trained on existing water segmentation datasets and tested for generalization to the ROSEBUD dataset. Results from further training show that modern semantic networks custom made for water recognition, and trained on marine images, can properly segment large areas, but they struggle to properly segment small obstacles in fluvial scenes without further training on the ROSEBUD dataset.https://www.mdpi.com/1424-8220/22/13/4681semantic segmentation training datasetunmanned surface vehicleobstacle detectiondeep learningcomputer vision |
| spellingShingle | Reeve Lambert Jalil Chavez-Galaviz Jianwen Li Nina Mahmoudian ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance Sensors semantic segmentation training dataset unmanned surface vehicle obstacle detection deep learning computer vision |
| title | ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance |
| title_full | ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance |
| title_fullStr | ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance |
| title_full_unstemmed | ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance |
| title_short | ROSEBUD: A Deep Fluvial Segmentation Dataset for Monocular Vision-Based River Navigation and Obstacle Avoidance |
| title_sort | rosebud a deep fluvial segmentation dataset for monocular vision based river navigation and obstacle avoidance |
| topic | semantic segmentation training dataset unmanned surface vehicle obstacle detection deep learning computer vision |
| url | https://www.mdpi.com/1424-8220/22/13/4681 |
| work_keys_str_mv | AT reevelambert rosebudadeepfluvialsegmentationdatasetformonocularvisionbasedrivernavigationandobstacleavoidance AT jalilchavezgalaviz rosebudadeepfluvialsegmentationdatasetformonocularvisionbasedrivernavigationandobstacleavoidance AT jianwenli rosebudadeepfluvialsegmentationdatasetformonocularvisionbasedrivernavigationandobstacleavoidance AT ninamahmoudian rosebudadeepfluvialsegmentationdatasetformonocularvisionbasedrivernavigationandobstacleavoidance |