Automatic Marine Debris Inspection
Plastic trash can be found anywhere, around the marina, beaches, and coastal areas in recent times. This study proposes a trash dataset called HAIDA and a trash detector that uses a YOLOv4-based object detection algorithm to monitor coastal trash pollution efficiently. Model selection, model evaluat...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Aerospace |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2226-4310/10/1/84 |
| _version_ | 1827629612399591424 |
|---|---|
| author | Yu-Hsien Liao Jih-Gau Juang |
| author_facet | Yu-Hsien Liao Jih-Gau Juang |
| author_sort | Yu-Hsien Liao |
| collection | DOAJ |
| description | Plastic trash can be found anywhere, around the marina, beaches, and coastal areas in recent times. This study proposes a trash dataset called HAIDA and a trash detector that uses a YOLOv4-based object detection algorithm to monitor coastal trash pollution efficiently. Model selection, model evaluation, and hyperparameter tuning were applied to obtain the best model for the lowest generalization error in the real world. Comparison of the state-of-the-art object detectors based on YOLOv3, YOLOv4, and Scaled-YOLOv4 that used hyperparameter tuning, the three-way holdout method, and k-fold cross-validation have been presented. An unmanned aerial vehicle (UAV) was also employed to detect trash in coastal areas using the proposed method. The performance on image classification was satisfactory. |
| first_indexed | 2024-03-09T13:54:52Z |
| format | Article |
| id | doaj.art-3f0507bf7851442e85661bb950f31cac |
| institution | Directory Open Access Journal |
| issn | 2226-4310 |
| language | English |
| last_indexed | 2024-03-09T13:54:52Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Aerospace |
| spelling | doaj.art-3f0507bf7851442e85661bb950f31cac2023-11-30T20:44:09ZengMDPI AGAerospace2226-43102023-01-011018410.3390/aerospace10010084Automatic Marine Debris InspectionYu-Hsien Liao0Jih-Gau Juang1Dynacolor, Inc., Taipei 114064, TaiwanDepartment of Communications, Navigation and Control, National Taiwan Ocean University, Keelung 202301, TaiwanPlastic trash can be found anywhere, around the marina, beaches, and coastal areas in recent times. This study proposes a trash dataset called HAIDA and a trash detector that uses a YOLOv4-based object detection algorithm to monitor coastal trash pollution efficiently. Model selection, model evaluation, and hyperparameter tuning were applied to obtain the best model for the lowest generalization error in the real world. Comparison of the state-of-the-art object detectors based on YOLOv3, YOLOv4, and Scaled-YOLOv4 that used hyperparameter tuning, the three-way holdout method, and k-fold cross-validation have been presented. An unmanned aerial vehicle (UAV) was also employed to detect trash in coastal areas using the proposed method. The performance on image classification was satisfactory.https://www.mdpi.com/2226-4310/10/1/84object detectionconvolutional neural networkmodel selectionmodel evaluationhyperparameter tuningUAV |
| spellingShingle | Yu-Hsien Liao Jih-Gau Juang Automatic Marine Debris Inspection Aerospace object detection convolutional neural network model selection model evaluation hyperparameter tuning UAV |
| title | Automatic Marine Debris Inspection |
| title_full | Automatic Marine Debris Inspection |
| title_fullStr | Automatic Marine Debris Inspection |
| title_full_unstemmed | Automatic Marine Debris Inspection |
| title_short | Automatic Marine Debris Inspection |
| title_sort | automatic marine debris inspection |
| topic | object detection convolutional neural network model selection model evaluation hyperparameter tuning UAV |
| url | https://www.mdpi.com/2226-4310/10/1/84 |
| work_keys_str_mv | AT yuhsienliao automaticmarinedebrisinspection AT jihgaujuang automaticmarinedebrisinspection |