Two- and three-dimensional segmentation and measurement of particles in the analysis of microscopic digital images of biological samples
Billy Wiseman Loo Jr., Bahram Parvin, Stephen S. Rothman · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1996
The authors present a Hough transform based image segmentation algorithm for automated detection, counting, and measurement of particles in two- and three-dimensional microscopic digital image data sets. The algorithm has proven to be both sensitive and specific for the particles of interest, even in the presence of noise and blurring. We apply the algorithm to the problem of automated compilation of population statistics on the size and mass of zymogen granules in pancreatic acinar cells. We present results describing the performance of the algorithm on digital phantoms, and image data from conventional fluorescence and confocal microscopes.