Applied Soft Computing 23 (2014) 259–269 Contents lists available at ScienceDirect Applied Soft Computing j ourna l h o mepage: www.elsevier.com/locate/asoc SAR images analysis based on polarimetric signatures Marzena Bielecka ∗ , Stanisława Porzycka-Strzelczyk, Jacek Strzelczyk AGH University of Science and Technology, Faculty of Geology, Geophysics and Environmental Protection, Department of Geoinformatics and Applied Computer Science, Al. Mickiewicza 30, 30-059 Cracow, Poland a r t i c l e i n f o Article history: Received 24 April 2013 Received in revised form 27 February 2014 Accepted 9 June 2014 Available online 17 June 2014 Keywords: Polarimetric signature Kohonen neural network Pattern recognition a b s t r a c t In the presented paper a new method of identification of canonical coherent scatterers in the quad- polarimetric SAR data are presented. The proposed method is based on the analysis of polarimetric signatures. The observed signatures are compared with the polarimetric signatures of four canonical objects: trihedral, dihedral and helix – right and left which represent basic scattering mechanisms: sin- gle bounce, double bounce and helix scattering. The polarimetric matrices are treated as vectors in a unitary space with a scalar product that generates the norm. A recognized object is classified to one of the four coherent classes by a Kohonen network. It is not trained in an iteration process but its weights are adjusted according to the given patterns. The network classification is supported by rules. The obtained maps of pixels that represent canonical objects are compared with a map of coherent scatterers which was obtained by using the polarimetric entropy approach. The developed method of canonical coherent scat- terers identification based on the polarimetric signatures analysis allows us not only to identify precisely the canonical coherent scatterers but also to determine the type of scattering mechanism characteristic for each of them. Since the proposed method works on a single-look (non-averaged) SAR data, it does not cause any spatial nor spectral decrease of amount of information because averaging is not conducted. Moreover, the proposed method will enable us the identification of a type of scattering mechanism in the canonical coherent pixels. This is an improvement in comparison to the existing methods. The obtained results should be more precise because the full polarimetric information about the scatterers is used in the identification procedure. © 2014 Elsevier B.V. All rights reserved. Introduction Synthetic-aperture radar (SAR for abbreviation) is a type of a radar that applies a relative motion between an antenna and its target region to provide distinctive long-term coherent-signal variations. Over the past decade extensive research in the area of the segmentation and classification of polarimetric SAR data in the context of outdoor scene analysis have been conducted [10,23,30,32,33,40]. In general, the classification algorithms for polarimetric SAR images can be divided into three main classes [16] based on: • physical scattering mechanisms inherent in data [41], • statistical characteristics of data [22,40], • image processing techniques [15]. ∗ Corresponding author. Tel.: +48 12 6174758. E-mail address: bielecka@agh.edu.pl (M. Bielecka). The aim of the described work was to develop a new method of canonical scatterer identification in quad-polarimetric SAR data. In the presented paper, the canonical coherent scatterers are defined as the radar targets for which one dominant scattering mecha- nism is characteristic and the total power of the received signal is relatively high. These scatterers mainly correspond with artifi- cial man-made targets such as buildings, power lines or bridges. The canonical coherent scatterers are a subgroup of all coherent scatterers which, in general, can be represented by a mix of differ- ent scattering mechanisms. There still does not exist any method designed directly to the identification of canonical coherent scat- terers. However, a few identification algorithms of the coherent scatterers has been already proposed in the literature, which can be related to the proposed method. One of them is based on the values of signal to clutter ratio [39]. In this method the intensity of studied pixel is compared with the intensity of surrounding pixels (clutter). If the ratio is large i.e. greater than 15 dB, then the pixel is classified as coherent. The overestimation of the clutter intensity in the urban areas where coherent targets are located very close to each other is the biggest disadvantage of this method. The coherent scatterers can be also identified by using the multitemporal analysis http://dx.doi.org/10.1016/j.asoc.2014.06.013 1568-4946/© 2014 Elsevier B.V. All rights reserved.