Image Classification of Cyanobacteria Microcystis Aeruginosa in raw Water Samples in Curitiba's Region

摘要

Surveys studies carried out in different countries demonstrated that about 75% of lake water samples contain toxic cyanobacteria [4, 5]. The toxin that is must recurrent in Brazilian water bodies are the microcystin [6]. Microcystins are hepatotoxins produced by some species of cyanobacteria, and act by inflicting damage to cells from the liver and other organs [7]. Cases around the world have been reported over time, including health injuries by contact direct or not with contaminated water, as well livestock and humans deaths [4]. Due to its sanitary importance, the water that the population consumes needs regular monitoring in order to avoid contact and possible adverse effects. The objective of this work is to propose a computational approach for the identification of Microcystis aeruginosa from images, in raw water reservoirs, automatically, using artificial intelligence techniques. The identification was developed using convolutional neural networks (CNN), where two distinct model were compared. The obtained results demonstrated the effectiveness of the proposed approach to solve the problem of recognizing cyanobacterias.