Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Coarse-coded higher-order neural networks for PSRI object recognition.

L Spirkovska¹, M B Reid

¹NASA Ames Res. Center, Mountain View, CA.

IEEE Transactions on Neural Networks

|January 1, 1993

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

The Sick and the Weak: Neuropathies/Myopathies in the Critically Ill.

Physiological reviews·2015

Same author

A-Z of nutritional supplements: dietary supplements, sports nutrition foods and ergogenic aids for health and performance--part 26.

British journal of sports medicine·2011

Same author

Optical calculation of potential fields for robotic path planning.

Applied optics·2010

Same author

Effects and correction of magneto-optic spatial light modulator phase errors in an optical correlator.

Applied optics·2010

Same author

Mapping considerations for optimal binary correlation filters.

Applied optics·2010

Same author

Experimental verification of modified synthetic discriminant function filters for rotation invariance.

Applied optics·2010

Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013

Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011

Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011

Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011

Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011

Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011

See all related articles

A novel coarse coding technique enables neural networks to achieve position, scale, and rotation invariant (PSRI) object recognition. This method demonstrates efficient learning, distinguishing objects within large image fields in under ten training passes.

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Object recognition systems often struggle with variations in object position, scale, and rotation.
Developing robust invariant recognition capabilities is crucial for many AI applications.

Purpose of the Study:

To introduce and evaluate a coarse coding technique for achieving position, scale, and rotation invariance (PSRI) in object recognition.
To demonstrate the effectiveness and limitations of this technique using simulations.

Main Methods:

Utilized a third-order neural network architecture.
Employed a coarse coding strategy for feature representation.
Trained the network on a large input field (4096x4096 pixels).

Related Experiment Videos

Main Results:

The neural network successfully learned to distinguish between two objects.
Achieved invariance to translation, in-plane rotation, and scale.
Required less than ten passes through the training dataset for convergence.
Empirically determined the limitations of the coarse coding technique within the PSRI domain.

Conclusions:

Coarse coding is a viable technique for developing efficient PSRI object recognition systems.
The studied neural network approach demonstrates rapid learning capabilities.
Further research is needed to fully understand and overcome the limitations of coarse coding in complex recognition tasks.