Processing framework to support maritime surveillance applications based on optical remote sensing images

Abstract

Nowadays, maritime security faces many kinds of problems – environmental hazards, unlawful actions such like piracy, cargo theft, illegal border crossing etc. These challenges bring the situational awareness of this domain to a high level of importance. Optical satellite images, captured during cloudless weather conditions, are valuable source of information about situation at sea. On the one hand, modern very high resolution (VHR) optical sensors (e.g. WorldView family) taking images with spatial resolution higher than 0.5 m per pixel, enabling to perform object (ship) detection tasks. On the other hand, high resolution (HR) sensors like Landsat-8 and Sentinel-2A(B) are able to cover relatively large areas and are suitable for environment monitoring tasks. Developed at the German Remote Sensing Data Center (DFD), part of the German Aerospace Center (DLR), Maritime Security Lab Processing Framework is intended to support the operational maritime surveillance near real time (NRT) services based HR and VHR optical satellite data. The Framework supports automated request driven processing from different satellite missions provided by a network of different ground stations and service providers. Actionable information products are created in an automatic processing chain including image pre-processing, data transcription and GUI based interactive value adding and validation. The paper will focus on the overall architecture of the framework including workflow of data handling, the interfaces and components, needed to enable fast data access for operator analysis and supervision.