Generalized performance parameter for single-threshold detection systems
- PMID: 20234664
- DOI: 10.1364/AO.19.003606
Generalized performance parameter for single-threshold detection systems
Abstract
A new generalized performance parameter for single-threshold detection systems is investigated that provides a more consistent measure of system performance than the conventional signal-to-noise ratio (SNR). It is shown explicitly that this parameter decreases monotonically as the Bayes risk and probability of error increase, and that it reduces to the SNR under appropriate conditions. An example is presented in which this parameter is monotonic with system performance, but the SNR is not. The analysis, which utilizes normalizing transformations, is also useful in the direct calculation of the Bayes risk, probability of error, and sensitivity, as is considered in detail subsequently (P. R. Prucnal, Applied Optics19, 3606 (1980)).
Similar articles
-
Error-probability noise benefits in threshold neural signal detection.Neural Netw. 2009 Jul-Aug;22(5-6):697-706. doi: 10.1016/j.neunet.2009.06.044. Epub 2009 Jul 3. Neural Netw. 2009. PMID: 19628368
-
[Analysis and experimental verification of SNR for a far ultraviolet imaging spectrograph in 115-180 nm].Guang Pu Xue Yu Guang Pu Fen Xi. 2010 Nov;30(11):3156-60. Guang Pu Xue Yu Guang Pu Fen Xi. 2010. PMID: 21284204 Chinese.
-
Relationship between dipole parameter estimation errors and measurement conditions in magnetoencephalography.IEEE Trans Biomed Eng. 1993 Sep;40(9):919-24. doi: 10.1109/10.245613. IEEE Trans Biomed Eng. 1993. PMID: 8288283
-
A generalized power-law detection algorithm for humpback whale vocalizations.J Acoust Soc Am. 2012 Apr;131(4):2682-99. doi: 10.1121/1.3685790. J Acoust Soc Am. 2012. PMID: 22501048
-
Comparing methods of analyzing fMRI statistical parametric maps.Neuroimage. 2004 Jul;22(3):1203-13. doi: 10.1016/j.neuroimage.2004.03.030. Neuroimage. 2004. PMID: 15219592 Review.
Cited by
-
Multiplication noise in the human visual system at threshold: 2. Probit estimation of parameters.Biol Cybern. 1982;43(2):87-96. doi: 10.1007/BF00336971. Biol Cybern. 1982. PMID: 7059629