A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS
Object detection from waterborne imagery is an essential aspect in maritime traffic management, navigation safety and coastal security. Building efficient autonomous systems, which can take decisions in critical situations, requires an integration of complex object detectors in real time. Object det...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/11/9/1638 |
| _version_ | 1797579400676376576 |
|---|---|
| author | Bogdan Iancu Jesper Winsten Valentin Soloviev Johan Lilius |
| author_facet | Bogdan Iancu Jesper Winsten Valentin Soloviev Johan Lilius |
| author_sort | Bogdan Iancu |
| collection | DOAJ |
| description | Object detection from waterborne imagery is an essential aspect in maritime traffic management, navigation safety and coastal security. Building efficient autonomous systems, which can take decisions in critical situations, requires an integration of complex object detectors in real time. Object detectors trained on generic datasets often give unsatisfactory results in complex scenarios like the maritime environment, since only a fraction of their images contain maritime vessels. Publicly available domain-specific datasets are scarce, and they are limited in the number of images and annotations. Compared to object detection in applications such as autonomous vehicles, maritime vessel detection is considerably reduced in computer vision research. This creates a deficit in exhaustive benchmarking studies for maritime detection datasets. To bridge this gap, we relabel the ABOships dataset and benchmark a state-of-the-art center-based detector, Centernet, on the newly relabeled dataset, ABOships-PLUS. We explore its performance under different feature extractors, and investigate the effect of object size and inter-class variation on detection accuracy. The reported benchmarking illustrates that the ABOships-PLUS dataset is adequate to use in supervised domain adaptation. The experimental results show that Centernet with DLA (Deep Layer Aggregation) as a feature extractor achieved the highest accuracy in detecting maritime objects overall (with mean average precision 74.4%). |
| first_indexed | 2024-03-10T22:36:35Z |
| format | Article |
| id | doaj.art-d064bd36e6ba4fe188c54b2a79eea6c6 |
| institution | Directory Open Access Journal |
| issn | 2077-1312 |
| language | English |
| last_indexed | 2024-03-10T22:36:35Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj.art-d064bd36e6ba4fe188c54b2a79eea6c62023-11-19T11:25:20ZengMDPI AGJournal of Marine Science and Engineering2077-13122023-08-01119163810.3390/jmse11091638A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUSBogdan Iancu0Jesper Winsten1Valentin Soloviev2Johan Lilius3Faculty of Science and Engineering, Åbo Akademi University, 20500 Åbo, FinlandFaculty of Science and Engineering, Åbo Akademi University, 20500 Åbo, FinlandFaculty of Science and Engineering, Åbo Akademi University, 20500 Åbo, FinlandFaculty of Science and Engineering, Åbo Akademi University, 20500 Åbo, FinlandObject detection from waterborne imagery is an essential aspect in maritime traffic management, navigation safety and coastal security. Building efficient autonomous systems, which can take decisions in critical situations, requires an integration of complex object detectors in real time. Object detectors trained on generic datasets often give unsatisfactory results in complex scenarios like the maritime environment, since only a fraction of their images contain maritime vessels. Publicly available domain-specific datasets are scarce, and they are limited in the number of images and annotations. Compared to object detection in applications such as autonomous vehicles, maritime vessel detection is considerably reduced in computer vision research. This creates a deficit in exhaustive benchmarking studies for maritime detection datasets. To bridge this gap, we relabel the ABOships dataset and benchmark a state-of-the-art center-based detector, Centernet, on the newly relabeled dataset, ABOships-PLUS. We explore its performance under different feature extractors, and investigate the effect of object size and inter-class variation on detection accuracy. The reported benchmarking illustrates that the ABOships-PLUS dataset is adequate to use in supervised domain adaptation. The experimental results show that Centernet with DLA (Deep Layer Aggregation) as a feature extractor achieved the highest accuracy in detecting maritime objects overall (with mean average precision 74.4%).https://www.mdpi.com/2077-1312/11/9/1638maritime vessel datasetship detectiondataset relabelingbenchmarkingdeep learningsmart shipping |
| spellingShingle | Bogdan Iancu Jesper Winsten Valentin Soloviev Johan Lilius A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS Journal of Marine Science and Engineering maritime vessel dataset ship detection dataset relabeling benchmarking deep learning smart shipping |
| title | A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS |
| title_full | A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS |
| title_fullStr | A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS |
| title_full_unstemmed | A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS |
| title_short | A Benchmark for Maritime Object Detection with Centernet on an Improved Dataset, ABOships-PLUS |
| title_sort | benchmark for maritime object detection with centernet on an improved dataset aboships plus |
| topic | maritime vessel dataset ship detection dataset relabeling benchmarking deep learning smart shipping |
| url | https://www.mdpi.com/2077-1312/11/9/1638 |
| work_keys_str_mv | AT bogdaniancu abenchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT jesperwinsten abenchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT valentinsoloviev abenchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT johanlilius abenchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT bogdaniancu benchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT jesperwinsten benchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT valentinsoloviev benchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus AT johanlilius benchmarkformaritimeobjectdetectionwithcenternetonanimproveddatasetaboshipsplus |