New Method for Epidermal Cells Detection in DIC images from Arabidopsis thaliana Leaves

Arabidopsis thaliana is the most widely used plant model for genetic analysis. Similarly, for cell cycle research, this model plant has been extensively used given the possibility of mutants phenotypic and molecular characterization in a straight-forward manner. In this sense, the first pair of extended leaves and their cellular and morphological changes in a developmental context serve to understand the genetic control that regulates cell cycle progression. To characterize specific cell cycle mutants, their first pair of leaves are analyzed using a microscope attached to a tube to draw cells by hand. However, this process is tedious, inaccurate and highly inefficient. Although an imaging method for cells detection in Arabidopsis thaliana leaves was previously proposed, the technique did not always allow cell borders detection and therefore, manual editing was required. In order to make a more efficient and accurate analysis, the preliminary results were improved to develop a new technique for automatic cell detection, based on phase congruency, using DIC microscopy images. The obtained results after the application of this novel technique are here presented.