Abdelmalek Toumi

A retrieval system from inverse synthetic aperture radar images: Application to radar target recognition

This paper presents an approach to aircraft target recognition using Inverse Synthetic Aperture Radar (ISAR) images. The goal of this work is to develop a robust algorithm to add Automated Target Recognition (ATR) capabilities to extract efficient feature vectors and ensure a comprehensive recognition process. In the first part on this paper, we used a database of ISAR images reconstructed from anechoic chamber simulations in order to extract efficient feature vectors. In the second part of this work, we proposed a recognition architecture to perform recognition tasks and provide a human operator with useful information for target recognition tasks. Several kinds of descriptor can be used to acquire information about target characteristics from radar signals and images. Indeed, a number of methods are currently used in automatic target recognition; notably based on ISAR imaging. However, target characteristic extraction from radar echoes remains a difficult task, and the methods are generally specific to either aircraft or ship recognition. In this work, we describe our approach to designing faster, more effective retrieval systems and a comprehensive architecture. Our approach uses global feature vectors for both ship and aircraft recognition. Firstly, the global feature vectors are described by two types of descriptor which have shown to be efficient in radar target recognition. The first type defines target shape obtained by watershed transformation, facilitating interpretation for the human operator. The second type of vector descriptor is based on so-called polar signatures, obtained using the polar mapping procedure. The latter are highly discriminative and thus significantly improve and facilitate target recognition, leading to greater precision and accuracy. Secondly, we describe the retrieval architecture suitable for the second type of vector descriptor, which checks invariance in relation to target rotation and scale. It is, in itself, more efficient and processing time at the retrieval step is reduced and controlled by the human operator. Finally, in order to validate our proposed feature vectors and architecture, the Support Vector Machine (SVM) classifier will be implemented; the results of which will be tested and presented in the last section of this paper.

Using watersheds segmentation on ISAR image for automatic target recognition

This paper deals with the processing adopted for shape extraction from the ID-presentation (image) in radar automatic target recognition field. The goal is to provide helpful information to human operator for target recognition. However, extracting the target characteristics from a radar echoes is the rather difficult task. Hence, several kinds of radar signatures can be employed to acquire information about target [10, 11]. In this paper, we present one approach for retrieval system for target recognition bused on ISAR-images in radar experimentation field. Then, we propose efficient features that deals with target shape which are extracted using Watersheds transformation. Of course, the target shape gives a better human interpretation.

Automatic recognition of ISAR Images: Target shapes features extraction

This paper presents a part of research study dealing with the initialization of an information system on radar automatic target recognition (ATR) chain. In this framework, we propose a set of helpful information at different levels of the ATR chain for decision making. The methodology used is the knowledge discovery process from data (KDD process).we focused in this paper on the data preparation step, more precisely Speckle denoising, edge detection and computation of features vector. We present the results of a several speckle filters that are largely used by ISAR imaging scientists (Lee, Frost, Kuan...). Fourier Descriptors have been used as features of the objects. Various similarity measures have been used and compared for target recognition.

Target recognition using IFFT and MUSIC ISAR images

We present in this paper an approach to achieve the aircraft target recognition task using Inverse Synthetic Aperture Radar (ISAR) images reconstructed using IFFT (Inverse Fast Furrier Transform) and MUSIC2D (multiple signal characterization) methods. The first goal of this work is to propose efficient features for target recognition using a polar mapping procedure, to make a polar signature, with well-designed classifier. The second goal of this work is to compare the recognition rate using the both reconstruction methods. Finally, the obtained results are given and compared for the two types ISAR images in order to validate our proposed feature vectors and retrieval scheme.

Estimation of Water Quality Parameters Using the Regression Model with Fuzzy K-Means Clustering

the traditional methods in remote sensing used for monitoring and estimating pollutants are generally relied on the spectral response or scattering reflected from water. In this work, a new method has been proposed to find contaminants and determine the Water Quality Parameters (WQPs) based on theories of the texture analysis. Empirical statistical models have been developed to estimate and classify contaminants in the water. Gray Level Co-occurrence Matrix (GLCM) is used to estimate six texture parameters: contrast, correlation, energy, homogeneity, entropy and variance. These parameters are used to estimate the regression model with three WQPs. Finally, the fuzzy K-means clustering was used to generalize the water quality estimation on all segmented image. Using the in situ measurements and IKONOS data, the obtained results show that texture parameters and high resolution remote sensing able to monitor and predicate the distribution of WQPs in large rivers.

Prediction of water quality parameters from SAR Images by using multivariate and texture analysis models

Remote sensing is one of the most important tools for monitoring and assisting to estimate and predict Water Quality parameters (WQPs). The traditional methods used for monitoring pollutants are generally relied on optical images. In this paper, we present a new approach based on the Synthetic Aperture Radar (SAR) images which we used to map the region of interest and to estimate the WQPs. To achieve this estimation quality, the texture analysis is exploited to improve the regression models. These models are established and developed to estimate six common concerned water quality parameters from texture parameters extracted from Terra SAR-X data. In this purpose, the Gray Level Cooccurrence Matrix (GLCM) is used to estimate several regression models using six texture parameters such as contrast, correlation, energy, homogeneity, entropy and variance. For each predicted model, an accuracy value is computed from the probability value given by the regression analysis model of each parameter. In order to validate our approach, we have used tow dataset of water region for training and test process. To evaluate and validate the proposed model, we applied it on the training set. In the last stage, we used the fuzzy K-means clustering to generalize the water quality estimation on the whole of water region extracted from segmented Terra SAR-X image. Also, the obtained results showed that there are a good statistical correlation between the in situ water quality and Terra SAR-X data, and also demonstrated that the characteristics obtained by texture analysis are able to monitor and predicate the distribution of WQPs in large rivers with high accuracy.

Log-polar and polar image for recognition targets

We describe in this paper, data processing algorithms applied on radar image in order to extract feature descriptors and then to perform recognition task. Several kinds of descriptors can be used to acquire information about target characteristics from radar images such as ISAR (Inverse Synthetic Aperture Radar) images. This paper presents two types of vector descriptors extracted via two minds of transformed images so-called polar and log-polar images obtained respectively from the polar and log-polar mapping. In order to guarantee the invariance of some geometrical transformation, additional processing are proposed. In this paper, we present the polar and log-polar transformations and then the classification scheme adapted on correspondent polar and log-polar templates. In the classification step, log-polar and polar mapping results are compared using adapted classification scheme.

Deep Learning for target recognition from SAR images

This paper deals with the problematic of automatic target recognition (ATR) using Synthetic Aperture Radar (SAR) images. In this work, the Deep Learning (DL) architecture is proposed and applied in order to recognize military vehicles from SAR images. We propose mainly in this work the deep learning algorithms based on convolutional neural network architecture. In the second step and in order to optimize the convolution of DL steps, we propose to use a convolutional auto-encoder which may be better suited to image processing. Its use provides several areas of the best results in the presence of noise on shifted and truncated images. To validate our approach, some experimentation results are given and compared. The obtained results show that the proposed approach of DL achieves a height recognition accuracy of 93%.

Integration of passive and active microwave remote sensing to estimate water quality parameters

In this paper, a new method has been presented to estimate of the Water Quality Parameters (WQPs) such as Electrical Conductivity (EC), salinity and Total Dissolved Salt (TDS). This method is mainly depended on the thermal and SAR images as an active and passive input satellite data. A Small Perturbation Method (SPM) using the Elfouhaily spectrum has been used as a physical model to calculate the electromagnetic scattering by the river surface. This method is basically used in the inversion process to find the physical parameters of water. The inversion process is fallen into three steps: first step includes, the estimation of the reflectivity coefficients via SAR image; then, the second step, the polarization coefficient (in VV or HH polarizations) is used to find the water dielectric constant; finally, the Debye formulation is used to find the salinity value. In this study, the thermal band is used to obtain the temperature of the water surface which we used in our proposed method. Total Dissolved Salts (TDS) is estimated by calculation of the electrical conductivity (EC) which is directly derived from the salinity value. To validate our methodology, a dataset of in situ data is used and the comparison between the WQP estimated using the proposed method and those obtained from in situ data is achieved. The obtained results shows, that the using of thermal image is helped to estimate salinity ranges close to in situ, consequently, enhance the produced TDS. The results also show that the thermal-TSX is helpful to improve the estimation and mapping of WQP for the practical use in the study area with high accuracy.

Visual salient sift keypoints descriptors for automatic target recognition

This paper addresses the problem of automatic target recognition (ATR) using inverse synthetic aperture radar (ISAR) images. In this context, we propose a novel approach for feature extraction to describe precisely an aircraft target from ISAR images. In our approach, a visual attention model is adopted to separate the salient regions from the background. After that, the scale invariant feature transform (SIFT) method is used to extract the keypoints and their descriptors. Then, a local salient feature is built by considering only the keypoints located in the salient region. For the classification step, the support vector machines (SVM) classifier is adopted. To validate the proposed approach, ISAR images database which was collected from anechoic chamber is used.