A computer vision system for segmentation of foraminifera

• Grant number: 14/11440-4
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: October 01, 2014
• End date: February 28, 2015
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Carlos Hitoshi Morimoto
• Grantee: Carlos Eduardo Leão Elmadjian
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Foraminifera are predominantly marine microscopic organisms, displaying highly variable shape. With diverse scientific applications, they can be used to detect pollution and acidification in the marine environment, as well as radioactive dating of rocks and geological periods, or even to infer paleoclimatology and sedimentary basins with oil. Currently, the study of these organisms is performed manually, from collecting the silt to identifying each foraminifera species, before samples can be used for other purposes. The objective of this undergraduate scientific project, with a proposed duration of 12 months, is to develop a computer vision system in order to automate part of this job. In particular, a system geared toward the segmentation of microscopic images of foraminifera. The candidate is going to initially study the foundations of both Computer Vision and Image Processing fields, focusing on the problem of segmentation of objects. During his studies, the student is going to implement some of the basic algorithms, so he can become familiar with the development tools. Once mastered the techniques and tools, some more advanced methods are going to be researched. This project is a resubmission, which takes into account the suggestions from the reviewer's assessment. In particular, the previous project dealt with problems of segmentation and classification, and in the current version the project was dimensioned to give more focus to the problem of segmentation. This project also describes in a more detailed fashion the methodology that is going to be employed.