Conjugate gradient and approximate Newton methods for an optimal probabilistic neural network for food color classification

被引:27
作者
Chtioui, Y
Panigrahi, S
Marsh, R
机构
[1] N Dakota State Univ, Dept Agr & Biosyst Engn, Fargo, ND 58105 USA
[2] N Dakota State Univ, Dept Comp Sci, Fargo, ND 58105 USA
关键词
probabilistic neural network; back-propagation neural network; conjugate gradient; approximate Newton; french fry;
D O I
10.1117/1.601972
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
The probabilistic neural network (PNN) is based on the estimation of the probability density functions. The estimation of these density functions uses smoothing parameters that represent the width of the activation functions. A two-step numerical procedure is developed for the optimization of the smoothing parameters of the PNN: a rough optimization by the conjugate gradient method and a fine optimization by the approximate Newton method. The thrust is to compare the classification performances of the improved PNN and the standard back-propagation neural network (BPNN). Comparisons are performed on a food quality problem: french fry classification into three different color classes (light, normal, and dark). The optimized PNN correctly classifies 96.19% of the test data, whereas the BPNN classifies only 93.27% of the same data. Moreover, the PNN is more stable than the BPNN with regard to the random initialization. The optimized PNN requires 1464 s for training compared to only 71 s required by the BPNN. (C) 1998 Society of Photo-Optical instrumentation Engineers. [S0091-3286(98)01711-5].
引用
收藏
页码:3015 / 3023
页数:9
相关论文
共 27 条
  • [1] Acton F.S., 1990, Numerical Methods That Work
  • [2] [Anonymous], 1992, MULTIVARIATE DENSITY
  • [3] [Anonymous], 1988, SELF ORG ASS MEMORY
  • [4] [Anonymous], 2021, PATTERN RECOGN, DOI DOI 10.1201/9780138744274
  • [5] EVALUATION OF PATTERN CLASSIFIERS FOR FINGERPRINT AND OCR APPLICATIONS
    BLUE, JL
    CANDELA, GT
    GROTHER, PJ
    CHELLAPPA, R
    WILSON, CL
    [J]. PATTERN RECOGNITION, 1994, 27 (04) : 485 - 501
  • [6] Broomhead D. S., 1988, Complex Systems, V2, P321
  • [7] Reduction of the size of the learning data in a probabilistic neural network by hierarchical clustering. Application to the discrimination of seeds by artificial vision
    Chtioui, Y
    Bertrand, D
    Barba, D
    [J]. CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 1996, 35 (02) : 175 - 186
  • [8] DENNIS JE, 1989, NUMERICAL METHODS UN
  • [9] GOODMAN PH, 1996, USER MANUAL NEVPROP
  • [10] INCREASED RATES OF CONVERGENCE THROUGH LEARNING RATE ADAPTATION
    JACOBS, RA
    [J]. NEURAL NETWORKS, 1988, 1 (04) : 295 - 307