Real-time detection of elliptic shapes for automated object recognition and object tracking Christian Teutsch a and Dirk Berndt a and Erik Trostmann a and Michael Weber a a Fraunhofer Institute for Factory Operation and Automation, Sandtorstrasse 22, 39106 Magdeburg, Germany ABSTRACT The detection of varying 2D shapes is a recurrent task for Computer Vision applications, and camera based object recogni- tion has become a standard procedure. Due to the discrete nature of digital images and aliasing effects, shape recognition can be complicated. There are many existing algorithms that discuss the identification of circles and ellipses, but they are very often limited in flexibility or speed or require high quality input data. Our work considers the application of shape recognition for processes in industrial environments and, especially the automatization requires reliable and fast algorithms at the same time. We take a very practical look at the automated shape recognition for common industrial tasks and present a very fast novel approach for the detection of deformed shapes which are in the broadest sense elliptic. Furthermore, we consider the automated recognition of bacteria colonies and coded markers for both 3D object tracking and an automated camera calibration procedure. Keywords: real-time ellipse detection, shape recognition, shape classification 1. INTRODUCTION Image processing is an integral part of our everyday life. Digital cameras automatically correct the pictures they have taken and at the airport, face recognition is an often discussed application. Processing images for object recognition is widely used in the industry, too. Cameras have to detect debris and impurity to assure a constant workflow, and the corresponding algorithms realize an automated segmentation of several objects. These systems are fast and mostly generate reproducible results. Basically, the function is quite simple. Objects are segmented from the known background and forwarded to an evaluation algorithm. If the background is unknown, then the objects texture and shape are analyzed. But there are many other applications where only a few information about the object structure and its environment is available. This requires complex and mostly time-consuming matching algorithms, which is problematic for automation and industrial applications. Especially in this scope, robust and fast algorithms are needed, based on adequate hardware, constant environmental conditions and a defined set of objects that must be detected. (a) (b) (c) (d) Figure 1. Applications for the detection of ellipses and deformed elliptic shapes: variably grown colonies of bacteria (a) and (b), coded markers on a board for camera calibration procedures (c) and markers on a car for spatial motion tracking and analysis (d). Further author information: christian.teutsch@iff.fraunhofer.de, Telephone: +49 (0)391 40 90 239