Amaral A L, Ginoris Y P, Nicolau A, Coelho M A Z, Ferreira E C
Institute for Biotechnology and Bioengineering, Centre of Biological Engineering, Universidade do Minho, Campus de Gualtar, 4710-057, Braga, Portugal.
Anal Bioanal Chem. 2008 Jun;391(4):1321-5. doi: 10.1007/s00216-008-1845-y. Epub 2008 Mar 8.
Protozoa are considered good indicators of the treatment quality in activated sludge systems as they are sensitive to physical, chemical and operational processes. Therefore, it is possible to correlate the predominance of certain species or groups and several operational parameters of the plant. This work presents a semiautomatic image analysis procedure for the recognition of the stalked protozoa species most frequently found in wastewater treatment plants by determining the geometrical, morphological and signature data and subsequent processing by discriminant analysis and neural network techniques. Geometrical descriptors were found to be responsible for the best identification ability and the identification of the crucial Opercularia and Vorticella microstoma microorganisms provided some degree of confidence to establish their presence in wastewater treatment plants.
原生动物被认为是活性污泥系统中处理质量的良好指标，因为它们对物理、化学和运行过程敏感。因此，有可能将某些物种或类群的优势与工厂的几个运行参数联系起来。这项工作提出了一种半自动图像分析程序，通过确定几何、形态和特征数据，并随后采用判别分析和神经网络技术进行处理，来识别污水处理厂中最常见的有柄原生动物物种。发现几何描述符具有最佳的识别能力，对关键的有盖虫属和小口钟虫属微生物的识别为确定它们在污水处理厂中的存在提供了一定程度的信心。
