Particle Shape Analysis Using Digital Image Processing

Background

The influence of biomass particle shapes cannot be ignored in particle transportation, mixing and fluidization processes. Various particle shapes result in different particle surface areas, in which are heat and mass transfer processes (Guo et al., 2012). Rodriguez et al. (2013) have presented a review about different methods and techniques to determine the geometrical shape of the particles. The authors observed that there is no agreement on the usage of the descriptors and is not clear which descriptor is the best. A large scale shape classification has been a problem. In addition, the authors considered that image analysis is a promising tool; it presents advantages like low time consumption or repeatability.

When non-spherical solid particles are observed through a microscope, various methods can be used for their sizing, resulting in terms of an equivalent spherical particle. Projected images in microscopes (optical, scanning and transmission) are in two-dimensional and depend on the orientation of the particles (Turbitt-Daoust, Alliet, Kaye, & Matchett, 2000). Particles in a stable orientation tend to have a maximum area causing microscopic measurements larger values than those presented by other methods, i.e., when smaller particle sizes are discarded.