A Hierarchical Graph Neuron Scheme for Real-Time Pattern Recognition

@article{Nasution2008AHG,
  title={A Hierarchical Graph Neuron Scheme for Real-Time Pattern Recognition},
  author={Benny Benyamin Nasution and Asad I. Khan},
  journal={IEEE Transactions on Neural Networks},
  year={2008},
  volume={19},
  pages={212-229}
}
The hierarchical graph neuron (HGN) implements a single cycle memorization and recall operation through a novel algorithmic design. The HGN is an improvement on the already published original graph neuron (GN) algorithm. In this improved approach, it recognizes incomplete/noisy patterns. It also resolves the crosstalk problem, which is identified in the previous publications, within closely matched patterns. To accomplish this, the HGN links multiple GN networks for filtering noise and… 

Towards real time multidimensional Hierarchical Graph Neuron (mHGN)

TLDR
The Hierarchical Graph Neuron (HGN) is developed so, that it is not only suitable for scrutinizing 1D- or 2D-patterns; it can also work on multidimensional patterns, which makes it suitable for real-time applications.

Towards real time multidimensional Hierarchical Graph Neuron (mHGN)

  • Benny Benyamin Nasution
  • Computer Science
    2014 International Conference on Computer and Information Sciences (ICCOINS)
  • 2014
TLDR
The Hierarchical Graph Neuron (HGN) is developed so, that it is not only suitable for scrutinizing 1D- or 2D-patterns; it can also work on multidimensional patterns, which makes it suitable for real-time applications.

A divide-and-distribute approach to single-cycle learning HGN network for pattern recognition

  • A. AminAsad I. Khan
  • Computer Science
    2010 11th International Conference on Control Automation Robotics & Vision
  • 2010
TLDR
The approach extends the single-cycle pattern recognition capability of Hierarchical Graph Neuron for wireless sensor networks into the more generic framework of computational grids and shows better noisy pattern recognition accuracy when size of the clusters is adapted to pattern data.

Graph neuron and hierarchical graph neuron, novel approaches toward real time pattern recognition in wireless sensor networks

TLDR
To overcome the issues of crosstalk available in the GN algorithm, Hierarchical Graph Neuron (HGN) an extended model of GN is presented which not only promises to deliver accurate results but also can be used for diverse types of applications in a multidimensional domain.

Distributed Pattern Transformation-Invariant Recognition Scheme for Real-Time Sensory Applications

TLDR
The proposed Cellular weighted Graph Neuron scheme solves the invariant pattern recognition for the first time in a fully distributed manner within a predictable single cycle learning duration which allows the use of the scheme within resource-constrained networks such as wireless sensor networks for event-detection.

Single-Cycle Image Recognition Using an Adaptive Granularity Associative Memory Network

TLDR
This paper demonstrates that the single-cycle learning approach can be successfully applied to denser patterns, such as black and white images, and is able to load-balance the pattern recognition processes, irrespective of the granularity of the underlying computational network.

Holographic Graph Neuron: A Bioinspired Architecture for Pattern Processing

TLDR
This approach significantly improves the noise resistance of the HGN architecture, and enables a linear time search for an arbitrary subpattern, through the use of vector symbolic architectures.

Building Context Aware Network of Wireless Sensors Using a Novel Pattern Recognition Scheme Called Hierarchical Graph Neuron

TLDR
This paper introduces a novel method using a highly scalable and distributed associative memory technique, called Hierarchical Graph Neuron (HGN), which overcomes the issue of crosstalk available in the original GN algorithm, and thus not only promises to deliver accurate results, but also can be deployed for diverse types of applications in a multidimensional domain.

Real-Time Multi Oriented Ancient Script Recognition using Single Layer Hierarchical Graph Neuron (SLHGN)

TLDR
In this improved approach, the SLHGN architecture is not only suitable for scrutinizing incomplete patterns, it can also accommodate multi oriented patterns and can contribute to recognize ancient scripts that are multi oriented, yet the same, patterns.

Parallel Pattern Recognition Using a Single-Cycle Learning Approach within Wireless Sensor Networks

  • A. AminAsad I. Khan
  • Computer Science
    2008 Ninth International Conference on Parallel and Distributed Computing, Applications and Technologies
  • 2008
TLDR
A parallel associative memory-based pattern recognition algorithm known as distributed hierarchical graph neuron (DHGN), able to reduce computational loads by efficiently disseminates recognition processes throughout the network, suitable to be deployed in wireless sensor networks.
...

References

SHOWING 1-10 OF 36 REFERENCES

Implementing a Graph Neuron Array for Pattern Recognition Within Unstructured Wireless Sensor Networks

TLDR
It is shown that a simple virtual topology overlay would enable distributed applications requiring stable structured networks to be deployed over dynamic unstructured networks without any alteration.

Parallel pattern recognition computations within a wireless sensor network

  • Asad I. KhanP. Mihailescu
  • Computer Science
    Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004.
  • 2004
TLDR
The computational properties of a wireless sensor network have been investigated by implementing a fully distributed pattern recognition algorithm within the network and it is shown that the test object could successfully differentiate between its internal stress states resulting from the changes to its external loading conditions.

Novel dynamic structure neural network for optical character recognition

TLDR
A novel dynamic structure neural network (DSNN) and a learning algorithm for training DSNN, which is implemented in C++ with man-machine interactive procedures and tested on many cases with very promising results.

Multi-associative neural networks and their applications to learning and retrieving complex spatio-temporal sequences

  • Lipo Wang
  • Computer Science
    IEEE Trans. Syst. Man Cybern. Part B
  • 1999
TLDR
This work designs a general system that can learn and retrieve complex spatio-temporal sequences with any MANN, and presents an implementation using the MCPN to demonstrate the applicability of this general system.

A sequential dynamic heteroassociative memory for multistep pattern recognition and one-to-many association

TLDR
A model that can perform multi step pattern recognition without the need for a special learning algorithm, and with the capacity to learn more than two pattern series in the training set is proposed.

Bidirectional associative memories

  • B. Kosko
  • Computer Science
    IEEE Trans. Syst. Man Cybern.
  • 1988
TLDR
The author proves that every n-by-p matrix M is a bidirectionally stable heteroassociative content-addressable memory for both binary/bipolar and continuous neurons.

Improvements of Complex-Valued Hopfield Associative Memory by Using Generalized Projection Rules

  • D. Lee
  • Engineering
    IEEE Transactions on Neural Networks
  • 2006
TLDR
It is shown that the well-known projection rule proposed by Personnaz can be generalized to complex domain such that the weight matrix of the CVHAM can be designed by using a simple and effective method.

Neural networks and physical systems with emergent collective computational abilities.

  • J. Hopfield
  • Computer Science
    Proceedings of the National Academy of Sciences of the United States of America
  • 1982
TLDR
A model of a system having a large number of simple equivalent components, based on aspects of neurobiology but readily adapted to integrated circuits, produces a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size.

Enhanced FMAM based on empirical kernel map

TLDR
An enhanced FMAM (EFMAM) based on the empirical kernel map is proposed and can significantly improve the auto-FMAM with respect to the recognition accuracy under hybrid-noise and computational effort.

A Pyramidal Neural Network For Visual Pattern Recognition

TLDR
This paper applies PyraNet to determine gender from a facial image, and compares its performance on the standard facial recognition technology (FERET) database with three classifiers: The convolutional neural network (NN), the k-nearest neighbor (k-NN), and the support vector machine (SVM).