Paper
16 April 2008 Automated ship image acquisition
Author Affiliations +
Abstract
The experimental Automated Ship Image Acquisition System (ASIA) collects high-resolution ship photographs at a shore-based laboratory, with minimal human intervention. The system uses Automatic Identification System (AIS) data to direct a high-resolution SLR digital camera to ship targets and to identify the ships in the resulting photographs. The photo database is then searchable using the rich data fields from AIS, which include the name, type, call sign and various vessel identification numbers. The high-resolution images from ASIA are intended to provide information that can corroborate AIS reports (e.g., extract identification from the name on the hull) or provide information that has been omitted from the AIS reports (e.g., missing or incorrect hull dimensions, cargo, etc). Once assembled into a searchable image database, the images can be used for a wide variety of marine safety and security applications. This paper documents the author's experience with the practicality of composing photographs based on AIS reports alone, describing a number of ways in which this can go wrong, from errors in the AIS reports, to fixed and mobile obstructions and multiple ships in the shot. The frequency with which various errors occurred in automatically-composed photographs collected in Halifax harbour in winter time were determined by manual examination of the images. 45% of the images examined were considered of a quality sufficient to read identification markings, numbers and text off the entire ship. One of the main technical challenges for ASIA lies in automatically differentiating good and bad photographs, so that few bad ones would be shown to human users. Initial attempts at automatic photo rating showed 75% agreement with manual assessments.
© (2008) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
T. R. Hammond "Automated ship image acquisition", Proc. SPIE 6963, Unattended Ground, Sea, and Air Sensor Technologies and Applications X, 696319 (16 April 2008); https://doi.org/10.1117/12.777091
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Artificial intelligence

Cameras

Photography

Image quality

Imaging systems

Databases

Zoom lenses

RELATED CONTENT

Third and first party ground truth collection for auto...
Proceedings of SPIE (February 12 2007)
Motion parametric modeling for very low bitrate video coding
Proceedings of SPIE (January 09 1998)
The diffractionator
Proceedings of SPIE (October 13 1995)

Back to Top