Binding and segmentation of multiple objects through neural oscillators inhibited by contour information

Abstract. Temporal correlation of neuronal activity has been suggested as a criterion for multiple object recognition. In this work, a two-dimensional network of simplified Wilson-Cowan oscillators is used to manage the binding and segmentation problem of a visual scene according to the connectedness Gestalt criterion. Binding is achieved via original coupling terms that link excitatory units to both excitatory and inhibitory units of adjacent neurons. These local coupling terms are time independent, i.e., they do not require Hebbian learning during the simulations. Segmentation is realized by a two-layer processing of the visual image. The first layer extracts all object contours from the image by means of “retinal cells” with an “on-center” receptive field. Information on contour is used to selectively inhibit Wilson-Cowan oscillators in the second layer, thus realizing a strong separation among neurons in different objects. Accidental synchronism between oscillations in different objects is prevented with the use of a global inhibitor, i.e., a global neuron that computes the overall activity in the Wilson-Cowan network and sends back an inhibitory signal.Simulations performed in a 50×50 neural grid with 21 different visual scenes (containing up to eight objects + background) with random initial conditions demonstrate that the network can correctly segment objects in almost 100% of cases using a single set of parameters, i.e., without the need to adjust parameters from one visual scene to the next. The network is robust with reference to dynamical noise superimposed on oscillatory neurons. Moreover, the network can segment both black objects on white background and vice versa and is able to deal with the problem of “fragmentation.”The main limitation of the network is its sensitivity to static noise superimposed on the objects. Overcoming this problem requires implementation of more robust mechanisms for contour enhancement in the first layer in agreement with mechanisms actually realized in the visual cortex.

[1]  K. D. Singh,et al.  Magnetic field tomography of coherent thalamocortical 40-Hz oscillations in humans. , 1991, Proceedings of the National Academy of Sciences of the United States of America.

[2]  Reinhard Eckhorn,et al.  Parallel processing by a homogeneous group of coupled model neurons can enhance, reduce and generate signal correlations , 1997, Biological Cybernetics.

[3]  Hansel,et al.  Synchronization and computation in a chaotic neural network. , 1992, Physical review letters.

[4]  DeLiang Wang,et al.  Synchronization and desynchronization in a network of locally coupled Wilson-Cowan oscillators , 1996, IEEE Trans. Neural Networks.

[5]  Terrence J. Sejnowski,et al.  Inhibition synchronizes sparsely connected cortical neurons within and between columns in realistic network models , 1996, Journal of Computational Neuroscience.

[6]  W. Singer.  Synchronization of cortical activity and its putative role in information processing and learning. , 1993, Annual review of physiology.

[7]  Michael J. Hawken,et al.  Macaque VI neurons can signal ‘illusory’ contours , 1993, Nature.

[8]  DeLiang Wang,et al.  Emergent synchrony in locally coupled neural oscillators , 1995, IEEE Trans. Neural Networks.

[9]  Irving Biederman,et al.  Higher-level vision , 1990 .

[10]  U. Polat,et al.  The architecture of perceptual spatial interactions , 1994, Vision Research.

[11]  Alfred Nischwitz,et al.  Local lateral inhibition: a key to spike synchronization? , 1995, Biological Cybernetics.

[12]  T. Sejnowski,et al.  Simulations of cortical pyramidal neurons synchronized by inhibitory interneurons. , 1991, Journal of neurophysiology.

[13]  D. Burr.  Temporal summation of moving images by the human visual system , 1981, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[14]  H B Barlow,et al.  Single units and sensation: a neuron doctrine for perceptual psychology? , 1972, Perception.

[15]  H. Schuster,et al.  A model for neuronal oscillations in the visual cortex , 1990, Biological Cybernetics.

[16]  Peter König,et al.  Stimulus-Dependent Assembly Formation of Oscillatory Responses: I. Synchronization , 1991, Neural Computation.

[17]  D. Sagi,et al.  The Architecture of Perceptual Spatial , 1994 .

[18]  Reinhard Eckhorn,et al.  Neural mechanisms of scene segmentation: recordings from the visual cortex suggest basic circuits for linking field models , 1999, IEEE Trans. Neural Networks.

[19]  K. Koffka.  Principles Of Gestalt Psychology , 1936 .

[20]  Isabelle Mareschal,et al.  A cortical locus for the processing of contrast-defined contours , 1998, Nature Neuroscience.

[21]  Zhaoping Li,et al.  A Neural Model of Contour Integration in the Primary Visual Cortex , 1998, Neural Computation.

[22]  DeLiang Wang,et al.  Image Segmentation Based on Oscillatory Correlation , 1997, Neural Computation.

[23]  Desmond J. Higham,et al.  An Algorithmic Introduction to Numerical Simulation of Stochastic Differential Equations , 2001, SIAM Rev..

[24]  Deliang Wang,et al.  Global competition and local cooperation in a network of neural oscillators , 1995 .

[25]  DeLiang Wang,et al.  Motion segmentation based on motion/brightness integration and oscillatory correlation , 2000, IEEE Trans. Neural Networks Learn. Syst..

[26]  Bard Ermentrout,et al.  When inhibition not excitation synchronizes neural firing , 1994, Journal of Computational Neuroscience.

[27]  G. N. Milstein,et al.  Mean-Square Numerical Methods for Stochastic Differential Equations with Small Noises , 1997, SIAM J. Sci. Comput..

[28]  W Singer,et al.  Visual feature integration and the temporal correlation hypothesis. , 1995, Annual review of neuroscience.

[29]  Peter König,et al.  Stimulus-Dependent Assembly Formation of Oscillatory Responses: III. Learning , 1992, Neural Computation.

[30]  H. Schuster,et al.  A model for neuronal oscillations in the visual cortex , 1990, Biological Cybernetics.

[31]  P. Kloeden,et al.  Numerical Solution of Stochastic Differential Equations , 1992 .

[32]  Ch. von der Malsburg,et al.  A neural cocktail-party processor , 1986, Biological Cybernetics.

[33]  J. Bouyer,et al.  Ventral mesencephalic tegmentum (VMT) controls electrocortical beta rhythms and associated attentive behaviour in the cat , 1982, Behavioural Brain Research.

[34]  Joachim M. Buhmann,et al.  Sensory segmentation with coupled neural oscillators , 1992, Biological Cybernetics.

[35]  G Tononi,et al.  Modeling perceptual grouping and figure-ground segregation by means of active reentrant connections. , 1991, Proceedings of the National Academy of Sciences of the United States of America.

[36]  C. Gilbert,et al.  Improvement in visual sensitivity by changes in local context: Parallel studies in human observers and in V1 of alert monkeys , 1995, Neuron.

[37]  G. Edelman,et al.  Reentrant signaling among simulated neuronal groups leads to coherency in their oscillatory activity. , 1989, Proceedings of the National Academy of Sciences of the United States of America.

[38]  J. Lund,et al.  Intrinsic laminar lattice connections in primate visual cortex , 1983, The Journal of comparative neurology.