Object detection for various types of vessels using the YOLO algorithm
1
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Owing to the development of computer vision technology, much effort is being conducted to apply it in the maritime field. In this study, we developed a model that can detect various types of ships using object detection. Nine types of ship images were downloaded, and bounding box processing of the ships in the images was performed. Among the You Only Look Once (YOLO) model versions for object detection, YOLO v3 and YOLO v5s were used to train the training set, and predictions were made on the validation and testing sets. For the validation and testing sets, both models made good predictions. However, as some mispredictions occurred in the testing set, recommendations for these are given in the last section.
Abstract
Owing to the development of computer vision technology, much effort is being conducted to apply it in the maritime field. In this study, we developed a model that can detect various types of ships using object detection. Nine types of ship images were downloaded, and bounding box processing of the ships in the images was performed. Among the You Only Look Once (YOLO) model versions for object detection, YOLO v3 and YOLO v5s were used to train the training set, and predictions were made on the validation and testing sets. For the validation and testing sets, both models made good predictions. However, as some mispredictions occurred in the testing set, recommendations for these are given in the last section.
참고문헌 (21)
- Y. Dong, "Ship object detec-tion of remote sensing image based on visual attention", Remote Sensing, 13(16), pp.3192~ (2021)[DOI]
- Y. Wang, "Au-tomatic ship detection based on RetinaNet using multi-res-olution Gaofen-3 imagery", Remote Sensing, 11(5) (2019)[DOI]
- R. Yang, "A novel CNN-based detector for ship detection based on rotatable bounding box in SAR images", IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14, pp.1938~1958 (2021)[DOI]
- J. Li, "Ship detection in SAR images based on an improved faster R-CNN", 2017 SAR in Big Data Era: Models, Methods and Applications, BIGSAR-DATA, pp.1~6 (2017)
- Y. L. Chang, "Ship detection based on YOLOv2 for SAR imagery", Remote Sensing, 11(7) (2019)[DOI]
- Z. Hong, "Multi-scale ship detec-tion from SAR and optical imagery via a more accurate YOLOv3", IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14, pp.6083~6101 (2021)[DOI]
- Z. Shao, "Multi-scale object detection model for autonomous ship navigation in maritime environ-ment", Journal of Marine Science and Engineering, 10(11), pp.1783~ (2022)[DOI]
- S. J. Lee, "Image-based ship detection and classification for un-manned surface vehicle using real-time object detection neural networks", International Offshore and Polar Engi-neering Conference (2018)
- Z. Shao, "Saliency-aware convolution neural network for ship detection in sur-veillance video", IEEE Transactions on Circuits and Sys-tems for Video Technology, 30(3), pp.781~794 (2020)[DOI]
- H. Li, "Enhanced YOLO v3 tiny network for real-time ship detection from visual image", IEEE Access, 9, pp.16692~16706 (2021)[DOI]
- J. H. Kim, "Object detection and classification based on YOLO-V5 with im-proved maritime dataset", Journal of Marine Science and Engineering, 10(3), pp.377~ (2022)[DOI]
- "Make Sense" (2024)[URL]
- "YAML" (2024)[URL]
- J. Redmon, "You only look once: Unified, real-time object detection", Pro-ceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp.779~788 (2016)
- A. Kuznetsova, (2020)
- A. Kuznetsova, "YOLOv5 ver-sus YOLOv3 for apple detection, Cyber-Physical Systems: Modelling and Intelligent Control", Studies in Systems, De-cision and Control, 338 (2021)[DOI]
- A. Khalfaoui, "Comparative study of YOLOv3 and YOLOv5’s performances for real-time person detection", 2022 2nd International Conference on Innovative Research in Applied Science, Engineering and Technology, pp.1~5 (2022)
- Y. Dai, "Efficient foreign object detection between PSDs and metro doors via deep neural networks", IEEE Access, 8, pp.46723~46734 (2020)[DOI]
- "YOLO v5" (2020)[URL]
- D. Dlužnevskij, "Inves-tigation of YOLOv5 efficiency in IPhone supported sys-tems", Baltic Journal of Modern Computing, 9(3), pp.333~344 (2021)[DOI]