Abstract
Ship type recognition has gained serious interest in applications required in the maritime sector. A large amount of the studies in literature focused on the use of images taken by shore cameras, radar images, and audio features. In the case of image-based recognition, a very large number and variety of ship images must be collected. In the case of audio-based recognition, systems may suffer from the background noise. In this study, we present a method, which uses the frequency domain characteristics with an image-based deep learning network. The method computes the fast Fourier transform of sound records of ships and generates the frequency vs magnitude graphs as images. Next, the images are given into the ResNet50 network for classification. A public dataset with nine different ship types is used to test the performance of the proposed method. According to the results, we obtained a 99% accuracy rate.
Sign-in/Register to access full text options 
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.
