A topological approach for detection of chessboard patterns for camera calibration - Emerging Trends in Image Processing, Computer Vision, and Pattern Recognition - page 436

CHAPTER 34

A topological approach for
detection of chessboard
paterns for camera calibration

Gustavo Teodoro Laureano 1 ; Maria Stela Veludo de Paiva 2 ; Anderson da Silva Soare s 1 ; Clarimar José Coelho 3

1 Computer

Science Institute (INF), Federal University of Goiás (UFG), Goiânia, Brazil
2 Engineering School of São Carlos (EESC), Electrical Engineering Department, University of São Paulo (USP), São Paulo, Brazil
3 Computer Science and Computer Engineering Department (CMP), Pontifical Catholic University of Goiás (PUC-GO), Goiânia, Brazil

Abstract
This work aims to automatically identify chessboard paterns for camera calibration. The method uses
a fast x-shaped corner detector and a geometric mesh to represent the relative association between fea-
tures. The mesh allows considering the regularity of the chessboard patern and a topological ilter is
presented. The matching between real world points and their image projections is done using neighbor-
ing properties in a filtered mesh. The point location is locally updated to the subpixel precision with a
specific x-corner detector. The calibration points are determined even when the patern is partially oc-
cluded. The experiments show that the proposed algorithm provides a robust detection of the chess-
board paterns and take great advantage of image frames. The method is applicable for both online and
of-line detection of chessboard paterns.

Keywords

Calibration paterns
Camera calibration
Patern recognition
Topological detection
Automatic camera calibration

1 Introduction
The camera calibration aims to determine the geometric parameters of the image formation pro-
cess [ 1 ] . This is a crucial step in many computer vision applications especially when metric in-
formation about the scene is required. In these applications, the camera is generally modeled