Abstract
Development of convolutional neural network (CNN) optimized for object detection, led to significant developments in ship detection. Although training data critically affect the performance of the CNN-based training model, previous studies focused mostly on enhancing the architecture of the training model. This study developed a sophisticated and automatic methodology to generate verified and robust training data by employing synthetic aperture radar (SAR) images and automatic identification system (AIS) data. The extraction of training data initiated from interpolating the discretely received AIS positions to the exact position of the ship at the time of image acquisition. The interpolation was conducted by applying a Kalman filter, followed by compensating the Doppler frequency shift. The bounding box for the ship was constructed tightly considering the installation of the AIS equipment and the exact size of the ship. From 18 Sentinel-1 SAR images using a completely automated procedure, 7489 training data were obtained, compared with a different set of training data from visual interpretation. The ship detection model trained using the automatic training data obtained 0.7713 of overall detection performance from 3 Sentinel-1 SAR images, which exceeded that of manual training data, evading the artificial structures of harbors and azimuth ambiguity ghost signals from detection.
Highlights
Surveillance over coastal regions was regarded as a crucial role in monitoring maritime traffic and in ensuring safety and securing resources [1]
The entire procedure of retrieving training data from synthetic aperture radar (SAR) image was comprised of four stages: (3.1) accurate interpolation of discrete automatic identification system (AIS) information on precise satellite time using the Kalman filter; (3.2) compensation of the Doppler frequency shift caused by the movement of each ship; (3.3) derivation of training data from predicted AIS position; and (3.4) ship detection by implementing the detection algorithm based on a modified conventional object detection algorithm
This research automated the entire procedure of extracting the bounding box from ships in SAR images by exploiting AIS information without intervention from artificial visual interpretation
Summary
Surveillance over coastal regions was regarded as a crucial role in monitoring maritime traffic and in ensuring safety and securing resources [1]. Within SAR images, the signatures of ships tended to have high backscattering coefficients, and they are discernable from background oceanic conditions [7]. A number of previous studies on conventional ship detection have focused on the implementation of such spectral characteristics, identifying ships from the ocean, which generally had low backscattering. Several studies further considered statistical approaches and the contrast between vessels and oceanic background: sampling the effective sub-band region from full bandwidth under the preprocessing stage [11], comparison between renowned spectral algorithms and methodologies applied on vessel detection [12], and principal component analysis (PCA) for ship context analysis on the identical type of warship-destroyer [13]. Discriminating vessels with their spectral characteristics caused a danger of inducing misdetection on oceanic features that resemble the SAR scattering of vessels such as sea clusters, surface waves, and artificial structures in inshore regions
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
