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

Automatic recognition of biological shapes using the Hotelling transform.

F J Sanchez-Marin1

  • 1Centro de Investigaciones en Optica, Loma del Bosque No. 115, Col. Lomas del Campestre, 37150 Leon, Guanajuato, Mexico. sanchez@foton.cio.mx

Computers in Biology and Medicine
|February 13, 2001
PubMed
Summary

Two novel contour-based methods enhance object recognition, overcoming translation, rotation, and scaling challenges. The Hotelling transform method significantly improves human corneal cell recognition compared to prior techniques.

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

[Impact of an on-line course on the level of knowledge of the OMI-AP program for primary care doctors].

Semergen·2014
Same author

Quantitative image quality analysis of a nonlinear spatio-temporal filter.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
Same author

Automatic recognition of biological shapes with and without representations of shape.

Artificial intelligence in medicine·2000
Same author

Automatic segmentation of contours of corneal cells.

Computers in biology and medicine·1999
Same author

The curvature function evolved in scale-space as a representation of biological shapes.

Computers in biology and medicine·1997

Area of Science:

  • Computer Vision
  • Biomedical Imaging

Background:

  • Object recognition systems often struggle with variations in translation, rotation, and scale.
  • Accurate shape representation is crucial for robust computerized object recognition.

Purpose of the Study:

  • To introduce and evaluate two novel contour-based techniques for object recognition that are invariant to translation, rotation, and scaling.
  • To compare the effectiveness of these new techniques against existing methods for recognizing biological cell structures.

Main Methods:

  • Developed a contour-based object recognition technique utilizing scale-space filtered coordinate functions and "largest diameters" of contours.
  • Developed a second contour-based technique employing the Hotelling transform on vector representations of contour points.
  • Applied both techniques to the specific task of recognizing human corneal endothelial cells within tissue samples.

Related Experiment Videos

Main Results:

  • The Hotelling transform technique demonstrated superior performance in recognizing human corneal endothelial cells.
  • This method showed considerable improvement over previously used representations like coordinate functions, curvature functions, and Fourier descriptors.
  • Both presented techniques successfully avoided complexities associated with translation, rotation, and scaling.

Conclusions:

  • Contour-based object recognition can be effectively achieved without explicit shape representation.
  • The Hotelling transform applied to contour vector representations offers a highly effective approach for cell recognition in biomedical imaging.
  • The proposed methods provide robust solutions for object recognition tasks with geometric variations.