Form Of Receiver Operating Characteristics Research Articles

Change detection methods for identification of mines in synthetic aperture radar imagery

This paper presents several methods for change detection in a pair of multi-temporal synthetic aperture radar (SAR) images of the same scene. Several techniques which vary in complexity were implemented and compared. Among the simple methods that were implemented are differencing, Euclidean distance, and image mean ratioing. These methods require minimal processing time, with little computational complexity, and incorporate no statistical information. These methods have demonstrated some degree of accuracy in detecting changes in SAR imagery. However, the presence of highly correlated speckle noise, misregistration errors, and nonlinear variations in SAR images motivated us to seek more sophisticated methods of change detection in order to obtain more favorable results. Therefore, methods were implemented which incorporated second order statistic calculations in making a change decision in efforts to mitigate false alarms arising from the aforementioned causes. Pre-whitened the data was created and then a Wiener prediction-based method, Euclidean distance measure and subspace projection method was implemented. The performance of these methods were compared using multi-look SAR images containing several targets (mines). The results are presented in the form of receiver operating characteristics (ROC) curves.

Optimal detection of crack echo families in elastic solids.

Optimal detection of a striplike crack residing in an isotropic elastic solid with coarse microstructure by means of ultrasonic nondestructive evaluation (NDE) is considered. A physics-based approach to derive an optimal detector, which achieves the theoretical limitations constrained by the underlying physics, is presented. State-of-the-art physical models of crack echoes and of stochastic backscattering from the material structure in elastic solids are introduced and unified with the theory of optimal detection to yield a practically useful nonlinear filter bank implementation of the optimal detector. Monte Carlo simulations of the detection performance for the special case of a striplike crack with uncertain angular orientation are presented in the form of receiver operating characteristics (ROCs). These new results represent the physical limitations for detecting a crack under the stated conditions and serve as performance bounds to which other detectors should be compared. A physics-based generalized likelihood ratio (GLR) detector, which relies on the same nonlinear filter bank as the optimal detector, is also presented for the special case of a striplike crack. A comparison between the optimal and the GLR detectors shows that the GLR detector only slightly reduces the performance.

Detection of Multiple Transient Signals With Unknown Arrival Times

The problem of optimal detection of signal transients with unknown arrival times contaminated by additive Gaussian noise is considered. The transients are assumed to be time continuous and belong to a parameterized family with the uncertainty about the parameters described by means of an a priori distribution. Under the assumption of a negligible probability that the independent transient observations overlap in time, a likelihood ratio is derived for the problem of detecting an unknown number of transients from the family, each transient with unknown arrival time. The uncertainty about the arrival times is assumed to be equal for all transients and is also described by means of a distribution. Numerical simulations of the performance of detecting a particular transient signal family are presented in the form of receiver operating characteristics (ROCs) for both the optimal detector and the classical generalized likelihood ratio test (GLRT). The results show that the optimal detector yields noticeable performance improvements over the GLRT. Moreover, the results show that the optimal detector may still outperform the GLRT when the true and modeled uncertainties about arrival times no longer agree.

An Analytical Series Expansion Solution to the Problem of Noncoherent Detection

The well-known noncoherent detection problem concerns optimal detection of an amplitude-modulated sinusoid, with an unknown phase angle, corrupted by additive Gaussian noise. The classical solution to this problem is the noncoherent detector which is known to be optimal if the envelope belongs to a specific set of functions or satisfies the narrow-band approximation i.e., that the bandwidth of the envelope is narrow in comparison with the (carrier) frequency of the sinusoid. In this work, an analytical series expansion solution to the likelihood ratio for the noncoherent detection problem is derived. This solution offers a generalization of the noncoherent detector in which the conditions imposed on the envelope stated above have been relaxed. Analytical expressions for the joint probability density functions (pdfs) of the in-phase and quadrature components, jointly expressed in polar coordinates, are also derived under the signal-plus-noise and the noise-only hypotheses, respectively. Numerical simulations of the detector performance are presented in the form of receiver operating characteristics (ROC) and minimum probability of error curves. The results from a comparison of the general analytical solution with the classical noncoherent detector show significant differences between the two detectors when the narrow-band approximation does not hold.

Performance Analysis of a Glrt in Late-Time Radar Target Detection - Abstract

This paper addresses the issue of target detection in Gaussian noise for wideband radar systems. Through the use of the parsimonious singularity expansion method (SEM) representation of the late-time transient radar return and Bayesian detection theory, an aspect independent detection method based on a generalized likelihood ratio test (GLRT) is developed. The resulting decision rule has a non-central Χ2 distribution which is used to analyze the performance of the GLRT in terms of the probability of detection (PD) and the probability of false alarm (PF) . Results illustrating the performance of the GLRT for two different thin wire targets are displayed in the form of receiver operating characteristics. Simulation results are generated to verify the analytical receiver operating characteristics.

Enhanced signal interception in the presence of interference

We consider the problem of intercepting a direct sequence spread spectrum waveform which is embedded in both narrowband interference and additive white Gaussian noise. The receiver we analyze is a broadband radiometer which has been enhanced in the following manner. Prior to entering the square-law device in the radiometer, the received waveform is passed through a transform domain processor which serves as a notch filter to null out the narrowband interference before it can contribute unwanted energy to the output of the radiometer. A central-limit theorem is established for the receiver's output which forms the basis for analyzing the performance of the receiver. For various combinations of system parameters, numerical results are presented in the form of receiver operating characteristics. >

Some stimulus conditions and observer characteristics which affect the form of auditory receiver operating characteristics

Emmerich, Gray, Watson, and Tanis [Percept. Psychophys. 11, 65–72 (1972)] discuss some anomalies in the forms of receiver operating characteristics (ROCs) obtained in auditory signal detection experiments. In order to investigate these anomalies further, ROCs have been determined for two stimulus conditions: a standard diotic signal detection condition, and an increment-detection (pedestal) condition. A comparison of data obtained in these two conditions shows, as expected, that the increment-detection task yields ROCs with overall slopes closer to 1.0 on normal-normal coordinates. However, anomalies in the forms of the ROCs for the increment-detection task remain. Evidence will be presented to show that subjects have certain response proclivities (for example, a tendency to avoid making the same response twice in a row) which may act to distort the forms of ROCs. In addition, the likelihood that short-term sensory changes also influence the data will be considered. Some consequences of such effects for inferences drawn from signal detection experiments, and the use of signal detection measures, will also be discussed.