Automatic and real time recognition of microalgae by means of pigment signature and shape

Coltelli P., Barsanti L., Evangelista V., Frassanito A. M., Passarelli V., Gualtieri P.

Environmental and Occupational Health  Image analysis  Public Health  J.3 LIFE AND MEDICAL SCIENCES  Feature extraction  General Medicine  Microalgae recognition  Management  Environmental monitoring  Monitoring  Policy and Law  Environmental Chemistry 

Microalgae are unicellular photoautotrophic organisms that grow in any habitat such as fresh and salt water bodies, hot springs, ice, air, and in or on other organisms and substrates. Massive growth of microalgae may produce harmful effects on the marine and freshwater ecological environment and fishery resources. Therefore, rapid and accurate recognition and classification of microalgae is one of the most important issues in water resource management. In this paper, a new methodology for automatic and real time identification of microalgae by means of microscopy image analysis is presented. This methodology is based on segmentation, shape features extraction, and characteristic colour (i.e. pigment signature) determination. A classifier algorithm based on the minimum distance criterion was used for microalgae grouping according to the measured features. 96.6% accuracy from a set of 3423 images of 24 different microalgae representing the major algal phyla was achieved by this methodology.

Source: Environmental science--processes & impacts. 15 (2013): 1397–1410. doi:10.1039/C3EM00160A

Publisher: Royal Society of Chemistry, Cambridge, Regno Unito

Metrics