Abstract
In the article synthesis procedure of mathematical model features of convolution neural network (CNN) is described. In order to improve the generalization capability of the network the training set is generated by adding a distorted image with changing of CNN receptive fields. This fact differs given procedure from the known procedures. We propose the reduction algorithm of an extended training set and the synthesis algorithm of features for CNN with non-standard receptive fields. The experiments results of the developed algorithms were shown in the article in order to assess of generalization capability changes of the convolution neural network. The experiments were performed with the hardware platform of the “Mechatronics” stand (SPA “Android techniques”, Russia).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bengio, Y.: Learning deep architectures for AI. Found. Trends Mach. Learn. 2(1), 1–127 (2009)
Bengio, Y., LeCun, Y.: Scaling learning algorithms towards AI. In: Large Scale Kernel Machines, MIT Press (2007)
Bishop, C.M.: Neural Networks for Pattern Recognition, pp. 482. Great Clarendon Street, USA (1995)
Cruz, V., Cristobal, G., Michaux, T., Barquin, S.: Invariant image recognition using a multi-network neural model. In: Proceedings International Joint Conference Neural Networks, vol. II, pp. 17–21 (1989). Electronic Neurocomputers
Modified National Institute of Standards and Technology (MNIST) (30.01.2016). http://yann.lecun.com/exdb/mnist/
Nemkov, R.: Dynamical change of the perceiving properties of neural networks as training with noise and its impact on pattern recognition. In: Young Scientists—International Workshop on Trends in Information Processing (YSIP) (2014) (30.01.2016). http://ceur-ws.org/Vol-1145/paper4.pdf/
Nemkov., R., Mezentseva, O.: The use of convolutional neural networks with non-specific receptive fields. In: The 4th International Scientific Conference: Applied Natural Sciences, pp. 284–289. Novy Smokovec (2013)
NYU Object Recognition Benchmark (NORB) (30.01.2016). www.cs.nyu.edu/ylclab/data/norb-v1.0/
Nemkov, R.M., Mezentseva, O.S.: Dynamical change of the perceiving properties of convolutional neural networks and its impact on generalization. Neurocomputers: Development and Application, vol. 2, pp. 12–18 (2015)
Nemkov, R.M.: Synthesis method of mathematical model parameters of the convolutional neural network with extended training set (30.01.2016). http://www.science-education.ru/125-19867/
Simard, P., Steinkraus, D., Platt, J.C.: Best practices for convolutional neural networks applied to visual document analysis. In: Proceeding of the Seventh International Conference on Document Analysis and Recognition (ICDAR03), vol. 2, pp. 958–964 (2003)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Nemkov, R.M., Mezentseva, O.S., Mezentsev, D. (2016). Using of a Convolutional Neural Network with Changing Receptive Fields in the Tasks of Image Recognition. In: Abraham, A., Kovalev, S., Tarassov, V., Snášel, V. (eds) Proceedings of the First International Scientific Conference “Intelligent Information Technologies for Industry” (IITI’16). Advances in Intelligent Systems and Computing, vol 451. Springer, Cham. https://doi.org/10.1007/978-3-319-33816-3_2
Download citation
DOI: https://doi.org/10.1007/978-3-319-33816-3_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-33815-6
Online ISBN: 978-3-319-33816-3
eBook Packages: EngineeringEngineering (R0)