Towards underwater sustainability using ROV equipped with deep learning system
Wu, Y.-C.; Shih, P.-Y.; Chen, L.-P.; Wang, C.-C.; Samani, H. (2020). Towards underwater sustainability using ROV equipped with deep learning system, in: 2020 International Automatic Control Conference (CACS), 4-7 November 2020, Hsinchu, Taiwan . pp. 1-5. https://dx.doi.org/10.1109/cacs50047.2020.9289788
|
| Beschikbaar in | Auteurs |
|
Documenttype: Congresbijdrage
|
| Author keywords |
Underwater, Sustainability, Image Enhancement, YOLO, ROV, Deep Learning |
| Auteurs | | Top |
- Wu, Y.-C.
- Shih, P.-Y.
- Chen, L.-P.
|
|
|
| Abstract |
Underwater pollution is a long-term environmental problem. Remotely Operated Vehicle (ROV) could promote the solution of kinds of the problem by detecting environmental problems such as rubbish. The paper suggests a system based on deep-learning algorithms that can detect rubbish underwater using ROV. We built our own dataset of three different types of underwater trash for a training model based on YOLO Neural Network architectures for object detection. In this paper we used YOLOv4. The training images from our dataset apply several filters for noise reduction and image enhancement processing to improve the accuracy of the results. In our experiment, the ROV could capture the image and stream it to the computer for analysis, enhancement, and recognition. |
|
