Camouflaged Targets Research Articles

Into the woods: visual surveillance of noncooperative and camouflaged targets in complex outdoor settings

Autonomous video surveillance and monitoring of human subjects in video has a rich history. Many deployed systems are able to reliably track human motion in indoor and controlled outdoor environments, e.g., parking lots and university campuses. A challenging domain of vital military importance is the surveillance of noncooperative and camouflaged targets within cluttered outdoor settings. These situations require both sensitivity and a very wide field of view and, therefore, are a natural application of omnidirectional video. Fundamentally, target finding is a change detection problem. Detection of camouflaged and adversarial targets implies the need for extreme sensitivity. Unfortunately, blind change detection in woods and fields may lead to a high fraction of false alarms, since natural scene motion and lighting changes produce highly dynamic scenes. Naturally, this desire for high sensitivity leads to a direct tradeoff between miss detections and false alarms. This paper discusses the current state of the art in video-based target detection, including an analysis of background adaptation techniques. The primary focus of the paper is the Lehigh Omnidirectional Tracking System (LOTS) and its components. This includes adaptive multibackground modeling, quasi-connected components (a novel approach to spatio-temporal grouping), background subtraction analyses, and an overall system evaluation.

Signature strength metrics for camouflaged targets corresponding to human perceptual cues

The goal of this research is to find signature strength metrics that correspond closely to the distinctness of a target as perceived by humans, especially for camouflaged targets. We consider metrics that attempt to measure the strength of three perceptual cues: contrast, tex- ture differences, and boundary strength. The contrast cue is measured with first-order metrics, differences in texture are measured with a second-order metric, and boundary strength is measured by computing contrast along the target-background boundary. We discuss a psycho- physical experiment designed to generate quantitative measurements of perceived target distinctness for comparison with the target signature strength metrics. This experiment involves paired comparisons of image stimuli, each containing a single random target pattern embedded at a known location in a random background pattern. The data from the psy- chophysical experiment are compared with computed values of the tar- get signature strength metrics, and a second-order image texture metric was found to exhibit the strongest correlation with the human data. © 1998 Society of Photo-Optical Instrumentation Engineers. (S0091-3286(98)03402-3) Subject terms: target signature metrics; target discrimination; perceptual cues; texture; camouflage.

A neural system for automatic target learning and recognition applied to bare and camouflaged SAR targets

This paper presents a neural based approach to target learning and recognition in synthetic-aperture radar imagery. Targets consist of a variety of camouflaged and uncamouflaged military vehicles taken at different radar view and depression angles in both spotlight and stripmap radar collection modes. Results from a variety of recognition experiments are reported.

Detecting the unexpected in photointerpretation.

A methodology for measuring photointerpretation performance from a single glance at a novel scene is presented. Subjects attempted to detect the presence or absence of a target. object, specified in advance by the object's name, from a cued position in a 150-ms flash of a line drawing of a real-world scene. The cue, a dot, was presented immediately after the presentation of the scene, so that the subjects were uncertain as to the position of the cued object. Miss rates as a function of distance of the cued position from central fixation, target size, and degree of target camouflage were determined. The experiment also explored how these functions varied with objects that were in unexpected locations, such as a sofa floating in a street scene. Objects in these locations violated the usual constraints which characterize the organization of real-world scenes. Large, uncamouflaged targets in a normal relation to their context suffered only modestly from the effects of increasing distance from fixation. But targets that were small, camouflaged, or undergoing violations of the relational constraints suffered marked increases in miss rates when presented only 3 or 4 deg from central fixation. Since humans can assimilate visual information faster than the eye can move, the maximum rate at which a fixation can be made provides a limit–the saccadic barrier–of visual information processing. In addition to furnishing general guidelines for scene processing displays, the results of this experiment serve to restrict the breaking of the barrier-with the aid of high-speed, high-capacity display systems–to the extraction of the setting or gist of a scene. Individual targets, unless large, uncamouflaged, and in expected locations, require direct fixation.

Determining the detectability range of camouflaged targets.

This paper describes a quick economical method for determining the maximum range at which camouflaged targets can be detected. The method produces results which agree within 10% of those obtained by conventional psychophysical methods.