A novel set of bilateral and radial symmetry shape descriptor based on contour information

REVOLLO, NATALIA; DELRIEUX, C; GONZÁLEZ JOSÉ, ROLANDO

IET COMPUTER VISION

INST ENGINEERING TECHNOLOGY-IET

Año: 2017 vol. 11 p. 226 - 236

1751-9632

Form and shape descriptors are among the most useful features for object identification13 and recognition. Even though there exists a widely used set of shape descriptors and underlying14 computational methods for their evaluation, frequently they fail to distinguish among very similar15 objects that they appear very different to the human eye. We propose a new set of shape16 descriptors based on a finer determination of the object symmetry axes, and a more accurate17 estimation of the bilateral and radial symmetry. These descriptors were thoroughly tested using18 several synthetic and real objects with varying degrees of symmetry. The methods for axes19 estimation and symmetry descriptors extraction outperform the widespread shape descriptors in20 recognizing and identifying among very similar objects.