Influence of the minimum particle size in the study of shape parameters by digital image analysis
Abstract:
This work addresses the influence of the size (in pixels) of digitized particles on the values of their shape parameters obtained using image analysis techniques, and considers the minimum sizes that provide reliable results. The population studied corresponded to basaltic grains from a volcanic sand. The shape parameters evaluated were solidity (So), sphericity (S), circularity (Ci) and fractal dimension (FD). The particle characterization was carried out by digital image analysis. The methodology was: acquisition and segmentation of mineral particles; quantification of particle shape parameters; application of the minimum particle sizes described in the literature to the studied samples and estimation of their shape parameters; representation and interpretation of changes in shape measures. The procedures applied allowed 71 particles to be identified and quantified using threshold segmentation. The use of three minimum particle sizes (130, 150 and 200 pixels) showed that changes in the shape parameter values can be as high as 2.80 % variation. The use of a minimum particle size for shape analysis ensures that incorrect numerical values produced by digital limitations are not included.
Año de publicación:
2020
Keywords:
- shape parameters
- image analysis
Fuente:

Tipo de documento:
Article
Estado:
Acceso restringido
Áreas de conocimiento:
- Ciencia de materiales
- Visión por computadora
Áreas temáticas:
- Métodos informáticos especiales
- Ciencias de la computación
- Medicina y salud