Automatic target recognition (ATR) performance is a function of image quality and its representation in the signature model generation and used in the ATR training process. This paper reports ATR performance as a function of synthetic aperture radar (SAR) image quality parameters including clutter-to-noise ratio (CNR) and multiplicative noise ratio (MNR). Images with specified image quality values were produced by introducing controlled degradations to the MSTAR public release data. Two different families of ATR algorithms, the statistical model-based classifier of DeVore, et al., and optimal tradeoff synthetic discriminant function (OTSDF) are applied to those data. Target classification accuracy was measured as a function of CNR/MNR for both the training and test data, indicating sensitivity of performance to a priori knowledge of these particular image quality parameters. Confusion matrices are expanded to include target aspect bins, providing visibility into performance as a function of aspect angle.
Composite correlation filters have been demonstrated in many automatic target recognition (ATR) applications because of their ability for class recognition and distortion-tolerance with shift invariance. Both the optimal tradeoff synthetic discriminant function (OTSDF) filters and optimal tradeoff distance classifier correlation filter (OTDCCF) approaches use parameters to combine multiple characteristics. Usually a set of filters is grouped into a bank for recognizing multiple targets across multiple geometric distortions. We extend these approaches to use independent tradeoff parameters in the filter synthesis for each class and grouping bin to improve classification. A method for determining the extended parameters is presented. Test results using the public SAR imagery MSTAR database are shown.
The focus of this paper is a genetic algorithms based method
to automate the construction of local feature based composite class
models that capture the salient characteristics of configuration
variants of vehicle targets in SAR imagery and increase the
performance of SAR recognition systems. The recognition models are
based on quasi-invariant local features, SAR scattering center
locations and magnitudes. The approach uses an efficient SAR
recognition system as an evaluation function to determine the fitness
of candidate members of a genetic population of new models and
synthetically generates composite class models that are more similar
to existing configurations than those configurations are to each
other. Intuitively, specific features of models of versions A and B
of an object may not match, because they are outside of some
tolerance, while they may both match some synthetic version C that is
somewhere in the middle. Experimental recognition results are
presented in terms of receiver operating characteristic (ROC) curves
to show the improvements in SAR recognition performance utilizing
composite class models of configuration variants of MSTAR vehicle
targets.
Composite correlation filters have been demonstrated in many synthetic aperture radar (SAR) automatic target recognition (ATR) applications because of their ability for class discrimination and distortion-tolerance with shift invariance. For many implementations, a simple model of white noise spectral density has been used to reduce output noise variance. Substituting a colored noise model or real clutter imagery in synthesizing correlation filters has shown improved clutter rejection and target recognition. However, the spectral response of SAR imagery is quite different than other types of images. We demonstrate the use of SAR clutter imagery and a new model for SAR clutter estimation for use in maximum average correlation height (MACH) and optimal tradeoff distance classifier correlation filter (OTDCCF) approaches. Test results using the MSTAR database are shown.
The MACH and DCCF correlation filter algorithms are evaluated using the publicly released MSTAR data base. These algorithms can be used as a matching engine for automatic target recognition in SAR imagery. In practice, the required filters can be synthesized using model based signature predictions. In addition, the MACH and DCCF algorithms are optimized to be robust to variations (distortions) in the target's signature. Unlike Matched Filtering or other exhaustive template based methods, the proposed approach requires very few filters. The paper describes the theory of the algorithm, key practical advantages and details of test results on the public MSTAR data base.
Recent developments in optimal trade-off based composite correlation filter methods have improved the recognition and classification of an object over a range of image distortions. We extend the capability of the distance classifier correlation filter introduced by Mahalanobis et al by using he optimal trade-off between different correlation criteria. These correlation filters can be used for the automatic target cueing or recognition of synthetic aperture radar (SAR) images. In this paper we will present results of designing these distortion-tolerant filters with simulated SAR imagery and testing with simulated SAR target images inserted into real SAR backgrounds.
Recent developments in composite correlation filter methods have improved the recognition and classification of an object over a range of image distortions. These correlation filters can be used for the automatic target cueing or recognition images. These new filter methods can be optimized for different correlation criteria in order to improve the recognition capability of the filter. In this paper we will present results of designing these distortion- tolerant correlation filters with simulated SAR imagery and testing with real and simulated SAR targets.
For optical implementation of correlation filters, we must design filters that can be implemented on available spatial light modulators (SLMs). Previous versions of computer code MEDOF: minimum Euclidean distance optimal filter produced various correlation filters using a single training image. In the newest version of MEDOF, we can use a training set of images to generate composite filters that include system noise and device constraints. In this paper we present implementation issues and results using composite filters generated by MEDOF.
In designing correlation filters for implementation on cross-coupled spatial light modulators, we are faced with the task of selecting the optimum gain and angle. Usually, this is carried out via an iterative search. In this paper, we introduce a direct method for identifying the optimum gain and angle. Simple simulation examples are included to illustrate the basic idea.
For optical implementation of correlation filters, we must design filters that can be implemented on available spatial light modulators (SLMs). These SLMs are usually characterized by the fact that their amplitude and phase responses are cross-coupled and thus cannot be independently controlled. Recently, we have introduced optimal trade-off filters that can be implemented on such cross-coupled devices. In this paper we present simulation results using these filters.
The performance of shift-invariant distance classifiers based on correlation filters is evaluated. First, the effect of noise on a classifier designed to recognize synthetic aperture radar (SAR) is observed. Then, a 2-class ATR designed to recognize infrared images of actual targets is evaluated. The results attest to the ability of the distance classifier to tolerate distortions, and recognize targets in the presence of noise and clutter.
The conventional Synthetic Discriminant Function (SDF) filters are complex-valued and thus cannot be accommodated on spatial light modulators that can represent only a subset of all possible values in the complex plane. Here, we compare the performance of different SDF filters designed to satisfy certain device restrictions.
Even in the absence of input noise, there is no guarantee that correlation peaks resulting from some filters such as the binary phase-only filters (BPOFs) will be at the origin when the input is the reference object centered. Simulation results are included that show this peak shift is usually negligible.
Designing filters for use with optical correlators is really an exercise in trading one performance measure against another. In this critical review, we present several different situations where such a tradeoff is carried out. An informed understanding of this law of nature is important in making sure that our goals in optical pattern recognition are realistic.
A systematic procedure is presented for designing optimal correlation filters for implementation on deformable mirror devices (DMDs) exhibiting cross-coupled amplitude and phase characteristics. The utility of the algorithm for designing such filters is illustrated using five different device characteristics: phase-only filter, a binary phase-only filter, a diagonal line characteristic, a DMD zeroth-order characteristic, and a DMD first-order characteristic. Results are also presented regarding the signal-to-noise ratio and peak-to-correlation energy obtainable using these filters. The performance achievable using DMD type characteristics was found to be close to that of phase-only filter.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.