Operating condition modeling for ATR fusion assessment

Abstract

Real world Operating Conditions (OCs) influence sensor data that in turn affects the performance of target detection and identification systems utilizing the collected information. The impact of operating conditions on collected data is widely accepted, but not fully characterized. OCs that affect data depend on sensor wavelength and associated scenario phenomenology, and can vary significantly between electro-optical (EO), infrared (IR), and radar sensors. This paper will discuss what operating conditions might be modeled for each sensor type and how they could affect automatic target recognition (ATR) systems designed to exploit their respective sensory data. The OCs are broken out into four categories; sensor, environment, target, and ATR algorithm training. These main categories will further contain subcategories with varying levels of influence. The purpose of this work is to develop an OC distribution model for the "real world" that can be used to realistically represent the performance of multiple ATR systems, and ultimately the decision made from the fused ATR results. An accurate OC model will greatly enhance the performance assessment of ATR and fusion systems by affording Bayesian conditioning in fusion performance analysis and aiding in the sensitivity analysis of fusion performance over different operational conditions. Accurate OC models will also be useful in the fusion algorithm operation.

Video image registration evaluation for a layered sensing environment

“…Lucas-Kanade Algorithm The Lucas-Kanade image alignment approach is unique compared to the other algorithms explored in this paper because it doesn't align images by selecting features and solving the correspondence problem [10][11][12]. Instead LK uses a gradient descent optimization technique, Gauss-Newton, to align images by making use of image intensity differences along with intensity gradient information [8][9][10].…”

“…Three techniques were developed under the Revolutionary Automatic Target Recognition and Sensor Research (RASER) grants and the fourth, the Lucas-Kanade (LK) algorithm, was adapted from [10][11][12] and tailored to the CLIF data set for registration comparison and evaluation. The correlation-based approach is intensity-based, the robust data alignment (RDA) approach is feature and intensity-based, the Scale Invariant Feature Transform (SIFT) is feature-based, and the LK is intensity-based.…”

Predicted radar/optical feature fusion gains for target identification

“…However; sensors can be adversely affected by environmental factors such as weather and time of day, imaging geometries which obscure the target of interest, and/or enemy counter measures. Sensor, target, and environmental variations make robust CID difficult [1]. Many recognition algorithms have been proposed and performance results published in the literature to address the combat identification problem for both moving and stationary targets.…”

Multisource information fusion for enhanced simultaneous tracking and recognition

“…The factors are collectively known as operating conditions (OCs). OCs can be divided into three major categories: sensor, target, and environment [1,2] . Many individual sensor OCs can be mitigated through sensor design or by altering collection geometry.…”