Abstract
An experiment is described which compares the performance of a neural network to human performance on a visual task which consists of detecting a target in a background image of correlated noise. A three-layer, feed-forward, multi-layer perceptron is trained to indicate the presence or absence of a target in images also presented to human observers. The basis for the comparison between the network and the human observers is the receiver operating characteristic (ROC) curve. Network performance is comparable to human performance for this particular task.
Similar content being viewed by others
References
Anderson TW (1984) An introduction to multivariate statistical analysis, 2nd edn. Wiley, New York
Baddeley AD, Colquhoun WP (1969) Signal probability and vigilance: a reappraisal of the signal-rate effect. Br J Psychol 60:169–178
Berger JO (1986) Statistical decision theory and Bayesian analysis, 2nd edn. Springer, Berlin Heidelberg New York
Dorfman DD, Alf EJ Jr (1968) Maximum likelihood estimation of parameters of signal detection theory — a direct solution. Psychometrica 33:117–124
Egan JP (1975) Signal detection theory and ROC analysis. Academic Press, New York
Gorman RP, Sejnowski TJ (1988) Analysis of hidden units in a layered network trained to classify sonar targets. Neural Net 1:75–89
Green DB, Swets JA (1966) Signal detection theory and psychophysics. Wiley, New York
Harrington L (1988) Computer generation of fractional brownian noise. ERIM Technical Report 658301-1-T
Huang WY, Lippman RP (1987) Comparisons between neural net and conventional classifiers. IEEE First International Conference on Neural Networks, San Diego, Calif, July 24–28, IV:485–493
Johnson NL, Kotz S (1970) Continuous univariate distributions, vol I. Wiley, New York
Keller JM, Crownover RM, Chen RY (1987) Characteristics of natural scenes related to the fractal dimension. IEEE Trans Pattern Anal Mach Intell PAMI-9:621–627
Kohonen T, Barna G, Chrisley R (1988) Statistical pattern recognition with neural networks: benchmarking studies. IEEE First International Conference on Neural Networks, San Diego, Calif, July 24–28, I:61–68
Myers KS, Barett HH, Borgstrom MC, Patton DD, Seely GW (1985) Effect of noise correlation on detectability of disk signals in medical imaging. J Opt Soc Am A 2:1752–1759
Ogilvie JC, Creelman CD (1968) Maximum Likelihood estimation of ROC curve parameters. J Math Psychol 5:377–391
Rao CR (1973) Linear statistical inference and its applications, 2nd edn. Wiley, New York
Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representation by error propagation. In: Rumelhart DE, McClelland JL, and the PDP Research Group (eds) Parallel distributed processing, vol 1, chap 8. MIT Press, Cambridge, Mass
Schoeneman PH, Tucker LR (1967) A maximum likelihood solution for the method of successive intervals allowing for unequal stimulus dispersions. Psychometrica 32:403–417
Van Trees HL (1968) Detection, estimation, and modulation theory, part I. Wiley, New York
Vogl TP, Mangis JK, Rigler AK, Zink WT, Alkon DL (1988) Accelerating the convergence of the back-propagation method. Biol Cybern 59:257–263
Wagner RF, Brown DG (1985) Unified SNR analysis of medical imaging systems. Phys Med Biol 30:489–518
Wagner RF, Insana MF, Smith SW (1988) Fundamental correlation lengths of coherent speckle in medical ultrasonic images. IEEE Trans Ultrason Ferroelectr Frequency Control 35:34–44
Zink WT, Mangis JK, Vogl TP (1988) A neural network that classifies Noisy T's and C's: an experimental comparison with statistical classifiers. Presented at INNS 1st International Conference, Boston, Mass, September 7–12, 1988. Neural Net (submitted for publication)
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Mangis, J.K., Voas, R.B., Zink, W.T. et al. Performance comparison of a neural network with human observers on a visual target detection task. Biol. Cybern. 62, 185–191 (1990). https://doi.org/10.1007/BF00198093
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF00198093