Image processing for particle characterization

A highly efficient and automatic technique of particle characterization using digital image processing and analysis is presented. The process, centered on the selection of individual objects, is composed of five general steps: image acquisition, preprocessing, segmentation, extraction, and representation of the characteristic parameters. The implementation of this technique is done using the Khoros system. Synthetic images are used to test the segmentation algorithms and the form-factor process. The proposed technique can deal with several Limiting problems such as background heterogeneity, touching or partially overlapping particles and a wide range of particle sizes and shapes. The performance of the method is illustrated by application to zirconia powder and to organic adipic acid particles on images obtained by light optical microscopy. The implemented method produces a statistical determination of various morphologic parameters including area, diameter, eccentricity, and form factor of the particles. (C) Elsevier Science Inc., 1996.