Human face identification using invariant descriptions and a genetic algorithm

R. Pinto-Elías, J. H. Sossa-Azuela

Resultado de la investigación: Capítulo del libro/informe/acta de congresoContribución a la conferencia

1 Cita (Scopus)

Resumen

© Springer-Verlag Berlin Heidelberg 1998. A new method to automatically identify a human face onto a 2D gray level image is presented. The method uses an invariant description of the face and a genetic algorithm to accomplish the task. The features used are the first four translation, rotation and scale moment invariants proposed by Hu [1]. In a first step, an image possibly containing a face is first divided into small cells of fixed size of 5 × 5 pixels. For each cell, the ordinary moments are next computed. From these, the corresponding Hu’s invariants are then derived. Human face identification is thus accomplished by grouping individual cells using a genetic algorithm by fitting a specific cost function. This cost function corresponds to the invariant description of a human face given in terms of the detected image features.
Idioma originalInglés estadounidense
Título de la publicación alojadaHuman face identification using invariant descriptions and a genetic algorithm
Páginas293-302
Número de páginas262
ISBN (versión digital)3540649921, 9783540649922
EstadoPublicada - 1 ene 1998
EventoLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) -
Duración: 1 ene 2014 → …

Serie de la publicación

NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen1484
ISSN (versión impresa)0302-9743

Conferencia

ConferenciaLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Período1/01/14 → …

Huella Profundice en los temas de investigación de 'Human face identification using invariant descriptions and a genetic algorithm'. En conjunto forman una huella única.

  • Citar esto

    Pinto-Elías, R., & Sossa-Azuela, J. H. (1998). Human face identification using invariant descriptions and a genetic algorithm. En Human face identification using invariant descriptions and a genetic algorithm (pp. 293-302). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1484).