Assessment of a linear pyroelectric array sensor for profile classification

Abstract

This paper describes the development of linear pyroelectric array systems for classification of human, animal, and vehicle targets. The pyroelectric array is simulated to produce binary profiles of targets. The profiles are classified based on height to width ratio using Naïve Bayesian classifiers. Profile widths of targets can vary due to the speed of the target. Target speeds were calculated using two techniques; two array columns, and a tilted array. The profile width was modified by the calculated speeds to show an improvement in classification results.

Range and velocity independent classification of humans and animals using a profiling sensor

“…Since neither of the classes is to be given preference, the hyper-plane should be located equidistant from the nearest points from class 1 and class 2. The distance of any point x from the hyper-plane is given by (5). Further, w and w 0 can be scaled so that the distance of the hyper-plane from the nearest points in class 1 and class 2 is set to unity.…”

“…Such a system can be implemented either by extracting signal from parts of a conventional focal plane array or by designing a sensor with only a single column of detectors 5 . In this research, we used a LWIR focal plane array camera (FLIR A40 camera) as a profiling sensor.…”

“…Such systems, however, are effective only for surveillance in confined spaces, such as trails that are bound by hills or vegetation. For open area surveillance, passive profiling sensors (the detected radiation is generated by the object) are more effective 5 . In our previous work we simulated a profiling sensor using a LWIR camera 6,7 .…”

Advancing Profiling Sensors with a Wireless Approach

“…Since each sensing element records these events using a fixed sampling rate, the data acquired for a slow passing object produces longer binary strings than for a faster moving object [19–21]. For instance, a human would generally produce longer strings than a vehicle, since the vehicle is more likely to move at a higher velocity.…”

“…To attain the average velocity of the object over the entire range of the sensor array, every inter-sensor velocity of the array must be calculated, not merely the velocity between adjacent sensing elements. The formula stated in Equation (1) expresses a two dimensional weighted average of the velocities across the entire array, where i and j are the initial and final sensing elements of interest, respectively, v ij is the velocity among elements calculated by Equation (2), and M is the number of sensing elements in the array [19,21]. Notice that the velocity indicated by Equation (2) is calculated by dividing the distance over the time it took the passing object to move from sensing element i to sensing element j .…”