Real time identification of algal phyla

algae identificationAlgae are identified in variety of research fields and used as indicators of water quality.  Quantifying and identifying algae is often a laborious task and requires a high level of skill.  Although there are several automated techniques now available, commercial developments have been limited.  Several software techniques based on imaging have been previously presented, however they’re often limited to only a few algal types, so not applicable to field samples. 

This paper presents an innovative method which provides real time recognition of multiple algaes.  The software uses image segmentation, shape features i.e. contours, centroid spectrum calculations and pigmentation.  The set up uses relatively simple hardware and no sample processing or fixation.  Coltelli et al tested the method on both cultured strains and field samples with the method correctly identifing 96.6% of 24 different algal phyla from 3423 images.

This article would be of interest to anyone involved in algal identification, whether from lab based cultures or water samples from the field. 

Automatic and real time recognition of microalgae by means of pigment and shape
Primo Coltelli, Laura Barsanti, Evangelista, Anna Maria Frassanito, Vincenzo Passarelli and Paolo Gualtieri
DOI: 10.1039/C3EM00160A

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)