Timothy C. Edwards

Probability of identification of small hand-held objects for electro-optic forward-looking infrared systems

This paper describes research on the measurement of the 50% probability of identification cycle criteria (N50,V50) for a set of hand-held objects normally held or used in a single hand. These cycle criteria are used to calibrate the Night Vision Electronic Sensors Directorate (NVESD) target acquisition models. The target set consists of 12 objects, from innocuous to potentially lethal. Objects are imaged in the visible, midwave infrared (MWIR), and long-wave infrared (LWIR) spectrum at 12 different aspects. Two human perception experiments are performed. The first experiment simulates an incremental constriction of the imaging systems modulation transfer function (MTF). The N50, and V50 calibration criteria are measured from this perception experiment. The second experiment not only simulates an incremental constriction of the system MTF but also down samples the imagery to simulate the objects at various ranges. The N50 and V50 values are used in NVTherm 2002 and NVThermIP, respectively, to generate range prediction curves for both the LWIR and MWIR sensors. The range predictions from both NVTherm versions are then compared with the observer results from the second perception experiment. The comparison between the results of the second experiment and the model predictions provides a verification of measured cycle criteria.

Sampling criteria for sensor simulation

Electro-optic sensor simulation and sensor design require a common understanding of spatial sampling. As part of the development of a high fidelity sensor simulation, the U.S. Army Night Vision Electronic Sensors Directorate has developed a methodology that enables the simulation designer to minimize the spatial sampling rate of the input image based on the sensor parameters. Gabor information theory is used to define the limit imposed on the ability to simultaneously repre- sent a spatial function and its Fourier transform. This is then coupled with the necessary consequences of sampling the image to develop limits that ensure minimal error. Resolution requirements for synthetic presen- sor imagery can be developed using this formalism. These requirements are not based on the instantaneous field of view (IFOV) or on the detec- tor width alone but are based on the full sensor point spread function.

Identification of handheld objects for electro-optic/FLIR applications

This paper describes research on the determination of the fifty-percent probability of identification cycle criterion (N50) for two sets of handheld objects. The first set consists of 12 objects which are commonly held in a single hand. The second set consists of 10 objects commonly held in both hands. These sets consist of not only typical civilian handheld objects but also objects that are potentially lethal. A pistol, a cell phone, a rocket propelled grenade (RPG) launcher, and a broom are examples of the objects in these sets. The discrimination of these objects is an inherent part of homeland security, force protection, and also general population security. Objects were imaged from each set in the visible and mid-wave infrared (MWIR) spectrum. Various levels of blur are then applied to these images. These blurred images were then used in a forced choice perception experiment. Results were analyzed as a function of blur level and target size to give identification probability as a function of resolvable cycles on target. These results are applicable to handheld object target acquisition estimates for visible imaging systems and MWIR systems. This research provides guidance in the design and analysis of electro-optical systems and forward-looking infrared (FLIR) systems for use in homeland security, force protection, and also general population security.

NVESD time-limited search model

A summary of the development and impact of the Night Vision and Electronic Sensors Directorate (NVESD) Time-Limited Search (TLS) Model for target detection is presented. This model was developed to better represent the search behavior when an observer is placed under time-constrained conditions. The three primary components of the search process methodology are (1) the average detection time (based on characteristics of the image), (2) occurrence and time delay associated with false alarms, and (3) the time spent searching a Field-of-View (FOV) before moving on to another FOV. The results of four independent perception experiments served as the basis for this methodology. The experiments, which were conducted by NVESD, portrayed time limited search conditions for different sensor resolution and background clutter levels. The results of the experiments showed that these factors influence the search process and their impacts are represented within the components of the TLS methodology. The discussion presents the problems with the current model and details the constraints that must be understood to correctly apply the new model.

Virtual MRTD experiments

Virtual minimum resolvable temperature difference (MRTD) measurements have been performed on an infrared sensor simulation based on FLIR 92 input parameters. By using this simulation , it is possible to perform virtual laboratory experiments on simulated sensors. As part of the validation of this simulation, a series of MRTD experiments were conducted on simulated and real sensors. This paper describes the methodology for the sensor simulation. The experimental procedures for both real and simulated MRTD are presented followed by a comparison and analysis of the results. The utility of the simulation in assessing the performance of current and notional sensors is discussed.

Staring imager minimum resolvable temperature measurements beyond the sensor half sample rate

It is a well-known phenomenon that three- and four-bar patterns can sometimes be resolved with staring sensors even when the bar pattern frequency is beyond the half sample rate of the sensor. When performing a minimum resolvable temperature (MRT) test, for example, the modulation of the target bars can be significant even when the fundamental frequency of the four-bar target is higher than the half sample rate of the sensor. We show that the modulation of a four-bar target goes to zero at 0.6 times the sample rate of the sensor (1.2 times the half sample rate) if the sample phasing is optimized. If the bar pattern contains a larger number of bars, the modulation cutoff approaches the half sample rate of the sensor. Further, we illustrate that, when the sample phase is optimized during the MRT measurement, the MRT performance of a sampled imager can exceed the measured MRT performance of an analog sensor with the same system and component modulation transfer functions.

Target identification performance of infrared imager models as a function of blur and sampling

An experiment has been performed at the U.S. Armys Night Vision and Electronic Sensors Directorate to fully test these models. The experiment imagery is intended to test the bounds of the models under which various blur and sampling is representative of the sensor in the task of target identification. The perception experiment is compared to the estimates of performance given by the various models. The model results are then compared and contrasted.

Dragonfly directional sensor

Abstract. This paper discusses the concept and hardware development of an all fiber-based, solid state, coherent array directional sensor that can locate and track bright objects against a darker background. This sensor is not an imager. It relies on the inherent structure of the global fiber distribution. Methods for characterizing and calibrating hardware embodiments are also presented.

Comparison of ID performance using real and synthetic imagery

Recent experiments performed at the U.S. Army Night Vision and Electronic Sensors Directorate (NVESD) provide significant insight into the validation of synthetic imagery for use in human perception experiments. This paper documents the procedures and results of target identification (ID) experiments using real and synthetic thermal imagery. Real imagery representing notional first generation and advanced scanning sensor systems was obtained. Parameters derived from the sensor data were used to generate synthetic imagery using the NVESD Paint the Night simulation. Both image sets were then used in a target identification experiment with trained human observers. Perception test results were analyzed and compared with metrics derived from the imagery. Several parameters missing from the original truth data were found to correlate with differences in the perception data. Synthetic data were regenerated using these additional parameters. A subsequent perception experiment confirmed the importance of these parameters, and a good match was obtained between real and synthetic imagery. While the techniques used in this series of experiments do not constitute a definitive method for validating synthetic imagery, they point to some important observations on validation. The main observation is that both target and local background characteristics must be sufficiently specified in the truth data in order to obtain good agreement between synthetic and real data. The paper concludes with suggestions as to the level of detail necessary for truth data when using synthetic imagery in perception experiments.

Use of synthetic imagery in target detection model improvement

Perception tests have been performed by Night Vision and Electronic Sensors Directorate (NVESD) addressing the process of searching an image with the intent of detecting a target of military importance. The imagery used in the experiments was generated using NVESDs Paint-the-Night (PTN) thermal image simulation. The use of PTN simulation permits the same scene and target to be viewed with different sensor characteristics (such as resolution, noise and sampling). This allows the isolation of single variables in an experimental environment and the evaluations of their effect on probability of detection. Typical first and second generation FLIR sensor effects were applied to each of 100 synthetic images resulting in an experimental data set with identical thermal signatures and sensor fields of view. Experimental results are presented and the advantages of using synthetic imagery to evaluate differences in sensor resolution, noise, and other characteristics are discussed.