Building roof contour recognition in high-resolution images using complex Zernike moments

• Grant number: 12/11022-2
• Support Opportunities: Scholarships in Brazil - Master
• Start date: September 01, 2012
• End date: May 31, 2014
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Geodesy
• Principal Investigator: Aluir Porfírio Dal Poz
• Grantee: Renata Nagima Imada
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil

Abstract

The study and experimental evaluation of the problem of characterization and recognition of building roof contour shapes using the complex Zernike moments are proposed in this research. The method of Zernike moments for shape analysis is based on the area of object of interest and it has two attractive characteristics for the proposed image analysis problem: 1) Zernike moments are invariant under geometric transformations of rotation, translation, and scaling; 2) Zernike moments are not sensitive to deformation along the region contour. These properties motivated the proposition of a database containing sketches (or models) of possible appearances of building roof contours in a given scene, associating to each sketch a feature vector based on Zernike moments. The Zernike moments would be also used to describe the shape of objects obtained by segmenting an image. The capacity of the Zernike moments in discriminating different building shapes, and also in discriminating building shapes from non-building shapes, would be assessed by using a metric (for example, the Euclidian distance) that evaluates the distance between a Zernike characteristic vectors computed from a building segmented in an image and another Zernike characteristic vector stored in the building sketch database. Finally, two challenges can be identified for this research: 1) tuning of parameters of the Zernike moments in order to maximize the capacity of the proposed method in discriminating building and non-building shapes; 2) construction and validation of the building contour sketch database.

Academic Publications

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

IMADA, Renata Nagima. Reconhecimento de contorno de edifício em imagens de alta resolução usando os momentos complexos de Zernike. 2014. Master's Dissertation - Universidade Estadual Paulista (Unesp). Faculdade de Ciências e Tecnologia. Presidente Prudente Presidente Prudente.