Information-theoretic bounds on target recognition performance based on degraded image data

IEEE Transactions on Pattern Analysis and Machine Intelligence - Tập 24 Số 9 - Trang 1153-1166 - 2002
A. Jain1, P. Moulin2, M.I. Miller3, K. Ramchandran4
1Qualcomm, Inc., San Diego, CA, USA
2Beckman Institute, Coordinated Science Laboratory and ECE Department, University of Illinois, Urbana, IL, USA
3Johns Hopkins University, Baltimore, MD, USA
4Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA, USA

Tóm tắt

This paper derives bounds on the performance of statistical object recognition systems, wherein an image of a target is observed by a remote sensor. Detection and recognition problems are modeled as composite hypothesis testing problems involving nuisance parameters. We develop information-theoretic performance bounds on target recognition based on statistical models for sensors and data, and examine conditions under which these bounds are tight. In particular, we examine the validity of asymptotic approximations to probability of error in such imaging problems. Problems involving Gaussian, Poisson, and multiplicative noise, and random pixel deletions are considered, as well as least-favorable Gaussian clutter. A sixth application involving compressed sensor image data is considered in some detail. This study provides a systematic and computationally attractive framework for analytically characterizing target recognition performance under complicated, non-Gaussian models and optimizing system parameters.

Từ khóa

#Target recognition #Object recognition #Remote sensing #Testing #Probability #Gaussian noise #Image coding #Image sensors #Sensor phenomena and characterization #Performance analysis

Tài liệu tham khảo

proakis, 1995, Digital Communications 10.1109/97.803428 lopresto, 1997, Proc IEEE Data Compression Conf feller, 1971, An Introduction to Probability Theory and Its Applications 10.1109/34.615449 bucklew, 1990, Large Deviation Techniques in Decision Simulation and Estimation 1997, IEEE Trans Image Processing, 6 10.1007/978-1-4757-2341-0 berger, 1984, Bayesian Decision Theory 10.1109/18.54897 rockefellar, 1972, Convex Analysis lehmann, 1986, Testing Statistical Hypotheses, 10.1007/978-1-4757-1923-9 10.1109/34.391409 10.1117/1.601317 fukunaga, 1990, Statistical Pattern Recognition 10.1109/83.552104 10.1109/34.709572 10.1117/1.602323 10.1109/TCOM.1977.1093935 10.1109/TCOM.1977.1093858 mallat, 1998, A Wavelet Tour of Signal Processing jain, 1999, Image Reconginition from Compressed Data: Performance Analysis van trees, 1968, Detection Estimation and Modulation Theory 10.1109/ICIP.2000.899587 cover, 1991, Elements of Information Theory, 10.1002/0471200611 duda, 2001, Pattern Classification ali, 1966, A General Class of Coefficients of Divergence of One Distribution from Another, J Royal Statistics Soc, 28, 132 10.1007/978-1-4615-3626-0