Automatic Ship Detection on Multispectral and Thermal Infrared Aerial Images Using MACS-Mar Remote Sensing Platform

Abstract

The Modular Aerial Camera System (MACS) is a development platform for optical remote sensing concepts, algorithms and special environments. For Real- Time Services for Maritime Security (EMSec joint project) a new multi-sensor configuration MACS-Mar was realized. It consists of 4 co-aligned sensor heads in the visible RGB, near infrared (NIR, 700-950 nm), hyperspectral (HS, 450-900 nm) and thermal infrared (TIR, 7.5-14 um) spectral range, a mid-cost GNSS/INS system, a processing unit and two data links. On-board image projection, cropping of redundant data and compression enable the instant generation of direct-georeferenced high resolution image mosaics, automatic object detection, vectorization and annotation of floating objects on the water surface. The results were transmitted over a distance up to 50 km in real-time via narrow and broadband data links and were visualized in a maritime situation awareness system. For the automatic onboard detection of objects a segmentation and classification workflow based on RGB, NIR and TIR information was developed and tested in September 2016. The completeness of the object detection in the experiment resulted in 95 %, the correctness in 53 %. Mostly bright backwash of ships led to overdetection of the number of objects, further refinement using water homogeneity in the TIR, as implemented in the workflow, could not be carried out due to problems with the TIR sensor. To analyze the influence of high resolution TIR imagery and to reach the expected detection quality a further experiment was conducted in August 2017. Adding TIR images the completeness was increased to 98 % and the correctness to 74 %.