Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Distortion tolerant pattern recognition based on self-organizing feature extraction.

J Lampinen1, E Oja

  • 1Dept. of Inf. Technol., Lappeenranta Univ. of Technol.

IEEE Transactions on Neural Networks
|January 1, 1995
PubMed
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

Characterization of diffraction gratings in a rigorous domain with optical scatterometry: hierarchical neural-network model.

Applied optics·2008
Same author

Optical scatterometry of subwavelength diffraction gratings: neural-network approach.

Applied optics·2008
Same author

PicSOM-self-organizing image retrieval with MPEG-7 content descriptors.

IEEE transactions on neural networks·2008
Same author

Adaptive algorithm for blind separation from noisy time-varying mixtures.

Neural computation·2001
Same author

Bayesian approach for neural networks--review and case studies.

Neural networks : the official journal of the International Neural Network Society·2001
Same author

Independence: a new criterion for the analysis of the electromagnetic fields in the global brain?

Neural networks : the official journal of the International Neural Network Society·2001
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

This study introduces a neural network system for identifying 2D objects in images, even with distortions. Its unsupervised feature learning and efficient design enable robust pattern classification for applications like facial recognition.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Machine Learning

Background:

  • Object recognition in digital images often struggles with distortions.
  • Existing systems may require extensive pre-classified training data.

Purpose of the Study:

  • To propose a generic, modular neural network system for distortion-tolerant 2D object detection and classification.
  • To develop a system with efficient parallel implementation and unsupervised feature learning.

Main Methods:

  • A pipeline architecture using feedforward neural networks for gradual distortion tolerance.
  • Feature extraction via distortion-tolerant Gabor transformations.
  • Minimum distortion clustering using multilayer self-organizing maps.
  • Object classification using a supervised one-layer subspace network.

Related Experiment Videos

Main Results:

  • The system demonstrates effective feature extraction and pattern classification in a distortion-tolerant manner.
  • Unsupervised learning allows the use of large, unclassified training datasets.
  • The feature space shows sufficient resolution for classifying multiple object classes with significant distortions.
  • Human faces and parts were successfully used as test object classes.

Conclusions:

  • The proposed system offers an efficient and robust solution for 2D object recognition under distortion.
  • The unsupervised, modular approach simplifies the learning process and enhances data utilization.
  • The system shows promise for real-world applications requiring reliable object identification from complex imagery.