Neural Computing and Applications

, Volume 18, Issue 3, pp 249–260 | Cite as

Effects of the number of hidden nodes used in a structured-based neural network on the reliability of image classification

Original Article


A structured-based neural network (NN) with backpropagation through structure (BPTS) algorithm is conducted for image classification in organizing a large image database, which is a challenging problem under investigation. Many factors can affect the results of image classification. One of the most important factors is the architecture of a NN, which consists of input layer, hidden layer and output layer. In this study, only the numbers of nodes in hidden layer (hidden nodes) of a NN are considered. Other factors are kept unchanged. Two groups of experiments including 2,940 images in each group are used for the analysis. The assessment of the effects for the first group is carried out with features described by image intensities, and, the second group uses features described by wavelet coefficients. Experimental results demonstrate that the effects of the numbers of hidden nodes on the reliability of classification are significant and non-linear. When the number of hidden nodes is 17, the classification rate on training set is up to 95%, and arrives at 90% on the testing set. The results indicate that 17 is an appropriate choice for the number of hidden nodes for the image classification when a structured-based NN with BPTS algorithm is applied.


Hidden nodes Backpropagation through structure Image classification Neural network Features set 


  1. 1.
    Zou W, Li Y, Lo KC, Chi Z (2006) Improvement of image classification with wavelet and independent component analysis (ICA) based on a structured neural network. Proceedings of IEEE world congress on computational intelligence’2006 (WCCI’2006), Vancouver, pp 7680–7685Google Scholar
  2. 2.
    Zou W, Lo KC, Chi Z (2006) Structured-based neural network classification of images using wavelet coefficients. In: Proceedings of third international symposium on neural networks (ISNN). Lecture notes in computer science 3972: advances in Neural Networks-ISNN2006, part II Springer, Chengdu, pp 331–336Google Scholar
  3. 3.
    Giles CL, Gori M (1998) Adaptive processing of sequences and data structures. Springer, BerlinGoogle Scholar
  4. 4.
    Sperduti A, Starita A (1997) Supervised neural networks for classification of structures. IEEE Trans Neural Netw 8:714–735Google Scholar
  5. 5.
    Goller C, Kuchler A (1996) Learning task-dependent distributed representations by back-propagation through structure. In: Proceedings IEEE international conference nerual networks, pp 347–352Google Scholar
  6. 6.
    Cho SY, Chi ZR, Siu WC, Tsio AC (2003) An improved algorithm for learning long-term dependency problems in adaptive processing of data structures. IEEE Trans Neural Netw 14:781–793Google Scholar
  7. 7.
    Wanas N, Auda G, Kamel MS, Karray F (1998) On the optimal number of hidden nodes in a neural network. IEEE Can Conf Elect Comp Eng 2:918–921Google Scholar
  8. 8.
    Mirchandani G, Cao W (1989) On hidden nodes for neural nets. IEEE Trans Circuits Syst 36(5):661–664MathSciNetGoogle Scholar
  9. 9.
    Nocera P, Quelavoine R (1994) Diminishing the number of nodes in multi-layered neural networks. IEEE world congress on computational intelligence, pp 4421–4424Google Scholar
  10. 10.
    Gorman RP, Sejnowski TJ (1988) Analysis of hidden units in a layered network trained to classify sonar targets. Neural Netw 1:75–89Google Scholar
  11. 11.
    Rojas R (2003) Networks of width one are universal classifiers. In: Proceedings of the international joint conference on neural networks, pp 3124–3127Google Scholar
  12. 12.
    How many hidden units should I use? (accessed 29 Aug 2007)
  13. 13.
    Tsoi AC (1998) Adaptive processing of data structures: an expository overview and comments. Technical report. Faculty of Informatics, University of Wollongong, AustraliaGoogle Scholar
  14. 14.
    Tsoi AC, Hangenbucnher M (1999) Adaptive processing of data structures. Keynote Speech. In: Proceedings of third international conference on computational intelligence and multimedia applications (ICCIMA ’99), 2–2 (summary)Google Scholar
  15. 15.
    Frasconi P, Gori M, Sperduti A (1998) A general framework for adaptive processing of data structures. IEEE Trans Neural Netw 9:768–785Google Scholar
  16. 16.
    Cho S, Chi Z, Wang Z, Siu W (2003) An efficient learning algorithm for adaptive processing of data structure. Neural Process Lett 17:175–190MATHGoogle Scholar
  17. 17.
    Dummer G, Winton R (1986) An elementary guide to reliability, 3rd edn. Pergamon Press, New York, 47pGoogle Scholar
  18. 18.
    Snedecor GW, Cochran WG (1989) Statistical methods. Iowa State University press, Ames, 507pGoogle Scholar

Copyright information

© Springer-Verlag London Limited 2008

Authors and Affiliations

  1. 1.Department of Electronic and Information EngineeringThe Hong Kong Polytechnic UniversityKowloonHong Kong
  2. 2.Shenzhen Institute of Advanced TechnologyShenzhenChina
  3. 3.Department of Mathematics and ComputingThe University of Southern QueenslandToowoombaAustralia
  4. 4.University of Central FloridaOrlandoUSA

Personalised recommendations