Determining the Optimal Boundaries of Alpha-entropy Classification Zone for Dual Circular Polarimetric Data Using the Concept of Maximum Similarity

Nowadays, SAR imaging is a well-developed remote sensing technique for providing high spatial resolution images of the Earth’s surface which provides a vast amount of information for environmental monitoring. Fully polarimetric (FP) SAR systems alternately transmit two orthogonal polarizations and receive the response of the scatters to each of them by two antennas with orthogonal polarizations. Transmitting two interleaved electromagnetic waves requires doubling the pulse repetition frequency which implies immediately that the image swath must be only half of the width of a single-polarized or dual-polarized SAR. In order to achieve a better swath width, and coincidentally reduce average power requirements and simplify transmitting hardware, compact polarimetric (CP) systems have been proposed with the promise of being able to maintain many capabilities of fully polarimetric systems (Souyris et al., 2005). One of the most important CP configurations is dual circular polarimetric (DCP) mode.In order to extract the physical scattering mechanism (PSM) of targets using polarimetric data many classification methods have been presented. One of the most common such methods is H-α decomposition (Cloude and Pottier, 1998) that is proposed for FP data. Its principle relies on the analysis of eigenvalues and eigenvectors of the coherency matrix. The space of scattering entropy (H) and mean alpha angle (α) namely H-α plane is used to classify the polarimetric image into 8 canonical PSMs.In recent years two approaches have been proposed in order to find dual H-α classification zones for DCP data. (Guo et al., 2012) proposed an H-α classification space by mapping the points of each PSM from the original FP data into the space of H-α for CP data and subsequently (Zhang et al., 2014) proposed an H-α space on the basis of the distribution centers and densities of different PSMs. Experimental results showed that the classification accuracy of each PSM is improved compared with the results of Guo’s H-α space, however Zhang’s method is not well accurate and there are still overlaps between different PSMs. The results of Zhang’s method for H- α boundaries is highly dependent on the choice of data. For example, in one data it might exist a special class of plants that are dominant in the image and in another one another class might be dominant. So, the maximum distribution densities of these two images are different from each other. Furthermore, the specifications of different sensors are different. For example, the base noise of each sensor is different and entropy is dependent on this parameter. So, for each specific sensor its own optimum boundaries should be found.
According to the fact that fully polarimetric data contains maximum polarimetric information, the efforts of the researchers in this field is to achieve the nearest information from CP data to FP data. Therefore, in this research we have found the H-α boundaries of DCP data which maximize the total class agreement of classification results of the DCP and FP data for RADARSAT-2 sensor. Two images over San Francisco and Vancouver acquired by Radarsat-2 at C-band in quad polarization mode, with the image size being 1151×1776 and 1766×1558 respectively have been used for this study. In order to evaluate the ability of the proposed H-α zones in comparison with Zhang’s zones, Each experimental image is classified into eight PSMs. Confusion matrices have been achieved and the resultant mean agreements have been calculated. It has been shown that the proposed boundaries have increased the mean agreements of the results by 3%.
In order to extract the physical scattering mechanism (PSM) of targets using polarimetric data many classification methods have been presented. One of the most common such methods is Cloude–Pottier H-α decomposition that is proposed for FP data. Its principle relies on the analysis of eigenvalues and eigenvectors of the coherency matrix. Entropy and α-angle are two important parameters for the interpretation of fully polarimetric data which are extracted from this method. They indicate the randomness of the polarisation of the back scattered waves and the scattering mechanisms of the targets respectively. For fully polarimetric data an H-α classification space has been presented. This H-α classification space is devided by H and α borders and cllassifies 8 feasible PSM regions without the need for training data.
In recent years two approaches have been proposed in order to find dual H-α classification zones for DCP data. In 2012, Guo proposed an H-α classification space by mapping the points of each PSM from the original FP data into the space of H-α for DCP data and extract approximate borders. Subsequently, in 2014 Zhang proposed an H-α space on the basis of the distribution centers and densities of different PSMs. Experimental results showed that the classification accuracy of each PSM is improved compared with the results of Guo’s H-α space, however Zhang’s method is not well accurate and there are still overlaps between different PSMs. Both Zhang’s and Guo’s methods are not based on an optimization method. Therefore, they do not present optimum H-α borders for classification of DCP data. Furthermore, each sensor has its own specifications. One of which is the system noise floor which affects entropy borders for classification. Thus, it is important to find optimum H-α boundaries for each sensor separately.
In this paper we have proposed a novel approach for finding optimum H/α classification borders for DCP data. The optimum borders have been found in such a way to maximize the agreement of the H-α classification results of DCP data with the H-α classification results of FP data. ‘Mean class agreement’ is introduced and the borders which maximize this parameter have been found. The results of classification using the proposed borders have been compared with the rival method and the superiority of the proposed method has been revealed.