Name
Abstract | ||
|---|---|---|
Maritime surveillance for intruders/illegal activities requires monitoring of a large area of the coastline. This task being manually exhaustive, would benefit immensely by application of object detection techniques to surveillance videos. However, object detection models trained on general objects datasets cannot be expected to give best performance for this scenario as marine vessels are only a small subset of these huge datasets and also do not classify the specific type of sea vehicle. Hence, their benchmarks are not appropriate for maritime surveillance. Some studies have been done with applications of Convolutional Neural Networks (CNN) for ship/boat detection on private and publicly available sea vessels datasets. This paper presents a summary of the benchmarks so far and presents our experiments of the latest object detection techniques for combined marine vessels dataset. A survey of the currently available datasets is also given. Results of our experiments in terms of mean Average Precision (mAP) and Frames Per Second (FPS) are presented. |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1109/IVCNZ51579.2020.9290533 | 2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ) |
Keywords | DocType | ISSN |
Ship detection,Deep learning,Marine vessel identification,Object detection,Intruder detection | Conference | 2151-2191 |
ISBN | Citations | PageRank |
978-1-7281-8580-4 | 0 | 0.34 |
References | Authors | |
5 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Mrunalini Nalamati | 1 | 0 | 0.34 |
| Nabin Sharma | 2 | 132 | 11.55 |
| Muhammad Saqib | 3 | 9 | 3.37 |
| M. Blumenstein | 4 | 168 | 31.87 |