Learning an Artificial Neural Network for Discovering Combinations of Bit-Quads to Compute the Euler Characteristic of a 2-D Binary Image
Document title: Learning an Artificial Neural Network for Discovering Combinations of Bit-Quads to Compute the Euler Characteristic of a 2-D Binary Image
Journal: Computación y sistemas
Database:
System number: 000560635
ISSN: 1405-5546
Authors: 1
2
4
2
Institutions: 1Centro de Investigaciones en Óptica, México
2Instituto Politécnico Nacional, Centro de Investigación en Computación, México
3Instituto Tecnológico y de Estudios Superiores de Monterrey, México
4Instituto Politécnico Nacional, Centro de Investigación y de Estudios Avanzados, México
Year:
Season: Ene-Mar
Volumen: 26
Number: 1
Pages: 411-422
Country: México
Language: Inglés
English abstract The Image Analysis community has widely used so-called bit-quads to propose formulations for computing the Euler characteristic of a 2-D binary image. Reported works have manually proposed different combinations of bit-quads to provide one or more formulations to calculate this important topological feature. This paper empirically shows how an Artificial Neural Network can be trained to find an optimal combination of bit-quads to compute the Euler characteristic of any binary image. We present results with binary images of different complexities and sizes and compare them with state-of-the-art machine learning algorithms.
Keyword: Euler characteristic,
Bit-quads,
Holes,
Objects,
Artificial neural network
Full text: Texto completo (Ver HTML) Texto completo (Ver PDF)