You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.


Ship detection on ENVISAT ASAR data: results, limitations and perspectives

Guillaume Hajduch(1) and Vincent Kerbaol(1)

(1) BOOST Technologies, 135, rue Claude Chappe, 29200 Brest, France


Ship detection is a crucial application for global monitoring for environment and security. It permits to monitor traffic, fisheries, and to associate ships with oil discharge. Nevertheless, each of those applications has specific constraints in terms of spatial coverage and resolution, radiometric quality, and contrast between vessels and ocean. By choosing adapted modes, polarisation and processing level ENVISAT/ASAR gives access to a wide range of data which can be optimally processed for specific applications. This paper address specific ship detection algorithms which have been developped for various demanding applications and extensively tested on ASAR data.

First, typical ship detection schemes and strategy are presented. Past research efforts on automatic target detection in SAR imagery have clearly demonstrated that no single detection algorithm will produce satisfactory results with sufficient detection sensitivity and small false alarm rate. Thus, ship detection systems generally consist of four stages: preprocessing; land masking; prescreening; and discrimination. In 2004, the demonstration of a ship detection as a add-on of oil spill detection service using ENVISAT ASAR Wide Swath images was defined and implemented in close cooperation with CEDRE over the French Zone de protection Ecologique (ZPE) in the Mediterranean Sea with such a four stage decomposition

Second, advanced processing for ship detection are addressed. Those specific processing take advantage of the huge amount of information that can be retrieved from low level product. For example some false alarm may be discarded by detecting azimut ambiguities and short life cycle events may be rejected by examining their effects on successives looks. The interest of using such advanced processing is illustrated on some examples, including coastal regions.

Third, under favorable conditions, additionnal information on detected ship targets may be retrieved, allowing to implement some basic classification. Typical dicriminant parameters are ship lengths, speed, and radar cross section. A basic classification strategy is proposed.

Eventually, all those detection and classification algorithm need to be tested to identify the advantages and drawbacks of different approaches, to strengthen final system robustness. This was the scope the project DECLIMS, Detection and Classification of Marine Traffic from Space, which is conducted by the EC Joint Research Centre. Preliminary results will be presented including mapping of european scale marine traffic.


Full paper


  Higher level                 Last modified: 07.10.03