The application of compressive sensing (CS) for imaging has been extensively investigated and the underlying mathematical principles are well understood. The theory of CS is motivated by the sparse nature of real-world signals and images, and provides a framework in which high-resolution information can be recovered from low-resolution measurements. This, in turn, enables hardware concepts that require much fewer detectors than a conventional sensor. For infrared imagers there is a significant potential impact on the cost and footprint of the sensor. When smaller focal plane arrays (FPAs) to obtain large images are allowed, large formats FPAs are unnecessary. From a hardware standpoint, this benefit is independent of the actual level of compression and effective data rate reduction, which depend on the choice of codes and information recovery algorithm. Toward this end, we used a CS testbed for mid-wave infrared (MWIR) to experimentally show that information at high spatial resolution can be successfully recovered from measurements made with a small FPA. We describe the highly parallel and scalable CS architecture of the testbed, and its implementation using a reflective spatial light modulator and a focal plane array with variable pixel sizes. We also discuss the impact of real-world devices and the effect of sensor calibration that must be addressed in practice. Finally, we present preliminary results of image reconstruction, which demonstrate the testbed operation. These results experimentally confirm that high-resolution spatial information (for tasks such as imaging and target detection) can be successfully recovered from low-resolution measurements. We also discuss the potential system-level benefits of CS for infrared imaging, and some of the challenges that must be addressed in future infrared CS imagers designs.
A novel method is presented for optimization of quadratic correlation filters (QCFs) for shift-invariant target detection in imagery. The QCFs are quadratic classifiers that operate directly on the image data without feature extraction or segmentation. In this sense, the QCFs retain the main advantages of conventional linear correlation filters while offering significant improvements in other respects. For example, multiple correlators work in parallel to optimize jointly the QCF performance metric and produce a single combined output, which leads to considerable simplification of the postprocessing scheme. In addition, QCFs also yield better performance than their linear counterparts for comparable throughput requirements. The primary application considered is target detection in infrared imagery for surveillance applications. In the current approach, the class-separation metric is formulated as a Rayleigh quotient that is maximized by the QCF solution. It is shown that the proposed method results in considerable improvement in performance compared with a previously reported QCF design approach and many other detection techniques. The results of independent tests and evaluations at the U.S. Army's Night Vision Laboratory are also presented.
In target recognition applications of discriminant or classification analysis, each 'feature' is a result of a convolution of an imagery with a filter, which may be derived from a feature vector. It is important to use relatively few features.We analyze an optimal reduced-rank classifier under the two-class situation. Assuming each population is Gaussian and has zero mean, and the classes differ through the covariance matrices: E, and 2• The following matrix is considered:We show that the k eigenvectors of this matrix whose eigenvalues are most different from 1/2 offer the best rank k approximation to the maximum likelihood classifier. The matrix A and its eigenvectors have been introduced by Fukunaga and Koontz; hence this analysis gives a new interpretation of the well known Fukunaga-Koontz transform.The optimality that is promised in this method hold if the two populations are exactly Gaussian with the same means. To check the applicability of this approach to real data, an experiment is performed, in which several 'modern' classifiers were used on an Infrared ATR data. In these experiments, a reduced-rank classifier-Tuned Basis Functions-outperforms others.The competitive performance of the optimal reduced-rank quadratic classifier suggests that, at least for classification purposes, the imagery data behaves in a nearly-Gaussian fashion. Downloaded From: http://proceedings.spiedigitallibrary.org/ on 06/27/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx Proc. of SPIE Vol. 5094 67 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 06/27/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx Proc. of SPIE Vol. 5094 71 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 06/27/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.