Detection of simply shaped objects on stationary images of an underlying surface by an interpreter/operator and by a computer vision system

摘要

Mathematical expressions were obtained for the probabilities of Type I and Type II errors when an interpreter/operator and a computer vision system search for axially symmetric and extended objects superimposed on an underlying surface, as a function of the noise correlation function, the ratio of object size to the correlation radius of the background, and the signal-to-noise ratio. Based on the results of a comparison of the potential approaches for addressing issues relating to detection of these objects, we propose an integrated system designed to improve the decision-making reliability. (C) 2016 Optical Society of America.