Abstract

To increase the confidence that the probability of acquiring (detecting or recognizing) objects from thermal-vision images under normalized conditions is no more nor less than a definite level, it is proposed to make an expert estimate of this probability for images obtained when the distances to the objects are, respectively, more or less than the specified distance. A technique is proposed for calculating the confidence probability corresponding to a specified lower limit of the confidence interval for the probability of acquiring the objects.

© 2012 OSA

Estimating detection and identification probabilities in maritime target acquisition

Jonathan M. Nichols, Kyle P. Judd, Colin C. Olson, James R. Waterman, and James D. Nichols
Appl. Opt. 52(12) 2531-2545 (2013)

Fast and precise 6D pose estimation of textureless objects using the point cloud and gray image

Wang Pan, Feng Zhu, Yingming Hao, and Limin Zhang
Appl. Opt. 57(28) 8154-8165 (2018)

Resolvability of Objects from the Standpoint of Statistical Parameter Estimation*

Carl W. Helstrom
J. Opt. Soc. Am. 60(5) 659-666 (1970)

Probability of detecting a coherent optical signal in thermal noise

C. Bendjaballah and K. Hassan
J. Opt. Soc. Am. 73(12) 1840-1843 (1983)

Estimating the probability density of the scattering cross section from Rayleigh scattering experiments

Nicolas Hengartner, Lawrence Talbot, Ian Shepherd, and Peter Bickel
J. Opt. Soc. Am. A 12(6) 1316-1323 (1995)