Abstract
This paper presents an original modular neural network architecture whose modules are multilayer perceptrons. The modules’ inputs are external inputs or hidden layers of other modules, thereby allowing them to be connected in a general manner. Based on this flexible architecture, networks with high numbers of inputs and outputs can be elaborated and properly trained. A suitable application is image transformation, i.e. the transformation of many input pixels into as many output pixels. Some architectural variations are presented; first the localization over a fraction of the network of a specific transformation’s training, and then the merging of two input images into a single output image. As a case study, we use the modular network to model mental images and their transformations (mental rotation, mental assemblage). It should eventually prove to be a valuable tool for image processing applications, such as super-resolution, or for the elaboration of complex cognitive systems.
Similar content being viewed by others
References
Haykin S (1999) Neural networks, a comprehensive foundation, 2nd edn. Prentice-Hall, Upper Saddle River
Principe JC, Euliano NR, Lefebvre WC (2000) Neural and adaptive systems, fundamentals through simulations. Wiley, New York
Cooper LA (1976) Demonstration of a mental analog of an external rotation. Percept Psychophys 19:296–302
Cooper LA, Shepard RN (1973) Chronometric studies of the rotation of mental images. In: Visual information processing. Academic, New York, pp 75–176
Shepard RN, Cooper LA (1982) Mental images and their transformations. MIT Press, Cambridge
Shepard RN, Metzler J (1971) Mental rotation of three-dimensional objects. Science 171:701–703
Hashem S (1997) Optimal linear combinations of neural networks. Neural Netw 10:599–614
Osherson DN, Weinstein S, Stoli M (1990) Modular learning. In: Schwartz EL (Ed) Computational neuroscience. MIT Press, Cambridge, pp 369–377
Jacobs RA, Jordan MI, Nowlan SJ, Hinton GE (1991) Adaptive mixtures of local experts. Neural Comput 3:79–87
Jordan MI (1994) A statistical approach to decision tree modeling. In: Proc of the Sev Annu ACM Conf on Comput Learn Theo. ACM Press, New York
Chen K, Chi H (1999) A modular neural network architecture for pattern classification based on different feature sets. Int J Neural Syst 9(6):563–581
Dailey MN, Cottrell GW (1999) Organization of face and object recognition in modular neural network models. Neural Netw 12:1053–1073
LeCun Y (1993) Efficient learning and second-order methods, a tutorial at NIPS 93, Denver
Hinton GE (1989) Connectionist learning procedures. Artif Intell 40:185–234
Marquardt D (1963) An algorithm for least-squares estimation of nonlinear parameters. SIAM J Appl Math 11:431–441
Jones R (1999) Garbage collection, algorithms for automatic dynamic memory management. Wiley, New York
Fukushima K (1980) Neocognitron: a self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biol Cybern 36:193–202
Fukushima K, Miyake S, Ito T (1983) Neocognitron: a neural network model for a mechanism of visual pattern recognition. IEEE Trans Syst Man Cybern 13(5):826–834
Fukushima K (1988) Neocognitron: a hierarchical neural network capable of visual pattern recognition. Neural Netw 1:119–130
Grossberg S (2000) How hallucinations may arise from brain mechanisms of learning, attention, and volition. J Int Neuropsychol Soc 6:583–592
Grossberg S (2003) How does the cerebral cortex work? Development, learning, attention, and 3D vision by laminar circuits of visual cortex. Behav Cogn Neurosci Rev 2:47–76
Hwang JN, Tseng YH (1993) 3D motion estimation using single perspective sparse range data via surface reconstruction neural networks. IEEE Int Conf Neural Netw 3:1696–1701
Watt A (2000) 3D computer graphics. Addison-Wesley, Reading
Grossberg S (2001) Neural substrates of visual percepts, imagery, and hallucinations. Technical Report CAS/CNS-01-11
Pylyshyn Z (2003) Return of the mental image: are there really pictures in the brain? Trends Cognit Sci 7(3):113–118
Linsker R (1986) From basic network principles to neural architecture: emergence of spatial-opponent cells. Proc Natl Acad Sci USA 83:7508–7512
Linsker R (1988) Self-organization in a perceptual network. IEEE Comput 21(3):105–117
Perlovsky LI (2003) Integration of language and cognition at pre-conceptual level. KIMAS, Boston, pp 280–285
Perlovsky LI (2004) Integrating language and cognition. IEEE Connect Newsl IEEE Neural Netw Soc 2(2):8–13
Oh IS, Suen CY (2002) A class-modular feedforward neural network for handwriting recognition. Pattern Recognit 35:229–244
Waibel A (1989) Modular construction of time-delay neural networks for speech recognition. Neural Comput 1:39–46
Hackbarth H, Mantel J (1991) Modular connectionist structure for 100-word recognition. Int Joint Conf Neural Netw Seattle 2:845–849
Auda G, Kamel M (1998) Modular neural network classifiers: a comparative study. J Intell Robot Syst 21:117–129
Lin Z, Shum HY (2001) On the fundamental limits of reconstruction-based super-resolution algorithms. Proc IEEE Comput Vis Pattern Recognit 1(8–14):1171–1176
Baker S, Kanade T (2002) Limits on super-resolution and how to break them. IEEE Trans Pattern Anal Mach Intell 24(9):1167–1183
Abraham WC, Robins A (2005) Memory retention—the synaptic stability versus plasticity dilemma. Trends Neurosci 28(2):73–78
Acknowledgments
We wish to thank the reviewers and our colleagues Dr. Adnan Acan and Dr. Marifi Güler for their most valuable and inspiring comments.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Carcenac, M. A modular neural network applied to image transformation and mental images. Neural Comput & Applic 17, 549–568 (2008). https://doi.org/10.1007/s00521-007-0152-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-007-0152-4