Uso da visão computacional para medição de crescimento fúngico

• Ano de defesa: 2012
• Autor(a) principal: Colella, Julio Cesar Tocacelli
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual de Maringá; Brasil; UEM; Maringá, PR
• Palavras-chave em Português: Agroinformática; Visão computacional; Bioinformática; Informática aplicada à agricultura; Ciências Agrárias; Agronomia
• Link de acesso: http://repositorio.uem.br:8080/jspui/handle/1/1168

In this area of computer vision there are very few studies about bioinformatics or even directed to agrocomputing, therefore, it becomes almost mandatory the use of bibliographic material related to areas such as biology, medicine and physics, as an example. Concerning the study of the behavior of microorganisms, particularly pathogenic, "in vitro", or also called vegetative growth, there aren?t also many studies related to mycelial growth, what turns the present material into some kind of source for the already published works. The Bioinformatics is a line of research involving multidisciplinary aspects which arose from the moment that the use of computational tools for the analysis of genetic data, biochemical and molecular biology has started. The agrocomputing - coined term to refer to information technology applied to agriculture - has gained space and becoming more and more organized, while in the field of plantpathology computer technology has been frequently used in the epidemiology of plant diseases. Computer Vision Systems in non-contact measurements are especially useful when dealing with small objects with complex geometry, for which conventional instruments cannot be used. The main objective of this work is to measure biological events as fungal growth, for example, which are not able to be captured by natural vision, but through computational tools such as computer vision. For each fungus were set up tables, with the measurement performed by the traditional way, every 12 hours, with another measurement using computer vision by digital ruler every each hour and, also, a junction table of both ones, turning the computer measurements into 12-hour measurements. The method of computer vision proved to be appropriate for measuring fungal growth and is a technique for easy operation, with greater accuracy for measuring mycelial growth of fungus when compared to the methods of the ruler and digital caliper. There was algo found an overestimation of the area of mycelial colony when methods of rule and the digital caliper were used.