Comparing behavioral receiver operating characteristic curves to multidimensional matched filters

Human factors experiments can be used to test whether a sensor can improve operator performance for detecting or recognizing a target.(1) Although hum an factors experiments are of tremendous value, these tests are time consum ing and resource intensive. To reduce costs associated with collecting beha vioral data, a two-dimensional matched filter is proposed. The objective is to compare and contrast behavioral and matched filter receiver operating c haracteristic (ROC) plots to determine whether the matched filter technique is a good predictor of human performance. Five different background images (three infrared band images, a chromatic-fused image, and monochromatic-fu sed image) were used with, and without, a target (airplane) present. False alarm and target detection probabilities were computed and results were plo tted on an ROC curve. The matched filter ROC curves were then compared to b ehavioral ROC curves. Results showed that the matched filter ROC curves wer e similar to behavioral ROC curves with color fusion, long-wave infrared sh owing the highest sensitivity, and mid-wave and short-wave infrared scenes were significantly less sensitive (near chance). These results indicate tha t the matched filter analysis may be used to model human behavior. (C) 2001 Society of Photo-Optical Instrumentation Engineers.