Image Analysis Based Quantification of Cells in Suspension Cultures for Producing Somatic Embryos

Abstract

A machine vision system which used macroscopic images of carrot cell suspensions was developed. This system was used to non-invasively evaluate cell growth. Packed cell volume (PCV) of the suspension was estimated by two features extracted by image analysis using the B component of the RBG color image. One feature was the number of pixels which were extracted by thresholding the B component at an appropriate value. It represented projected area of cells and cell aggregates, and could be used for PCV estimation. Another feature was the sum total of B values of all pixels in the suspension image. It was also used for PCV estimation having the advantage that it did not require a procedure for extracting pixels corresponding to cells. Using this system, an increase in specific growth rate with a decrease of embryogenic potential was monitored in a carrot embryogenic cell suspension culture. This system showed the potential to be used to select cell lines with desirable growth characteristics for somatic embryo production.