Image segmentation based in superpixels and complex networks: an application to birds census

• Grant number: 11/05802-2
• Support Opportunities: Scholarships in Brazil - Doctorate
• Effective date (Start): July 01, 2011
• Effective date (End): October 31, 2013
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: João Do Espirito Santo Batista Neto
• Grantee: Glenda Michele Botelho
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The birds census is an important activity for identification of the reduction of the amount of individuals of a population, allowing better targeting of management activities of conservation programs. Typically, this activity is performed by land monitoring and, more recently, by air monitoring, using images. This project aims to define, enhance and develop image processing techniques for automatic segmentation of aerial images making it possible to count birds. The images used in this project have high resolution and the objects of interest (birds) occupy only a portion of the images. Therefore, the segmentation is performed in two stages: 1) pre-segmentation, to eliminate areas irrelevant and 2) final segmentation to estimate the amount of birds. Pre-segmentation will be performed by techniques of complex networks and the final segmentation is made by Markov Random Fields, due to its ability to represent microtextures (birds). It is expected to contribute to advancing the state of the art of the birds census by processing of aerial images, to enhance and develop methods capable of working with high resolution images and create approaches that deal with traditional segmentation algorithms reducing its computational cost.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

LINARES, OSCAR A. C.; BOTELHO, GLENDA MICHELE; RODRIGUES, FRANCISCO APARECIDO; BATISTA NETO, JOAO. Segmentation of large images based on super-pixels and community detection in graphs. IET IMAGE PROCESSING, v. 11, n. 12, p. 1219-1228, DEC 2017. (11/05802-2)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

BOTELHO, Glenda Michele. Image segmentation based on complex networks and superpixels: an application to birds census. 2014. Doctoral Thesis - Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB) São Carlos.