Fast Image Segmentation, Object Recognition and Localization in a RoboCup Scenario

作者: Thorsten Bandlow , Michael Klupsch , Robert Hanek , Thorsten Schmitt

DOI: 10.1007/3-540-45327-X_13

关键词: Machine visionArtificial intelligenceImage segmentationRobotComputer scienceComputer visionRoboticsImage processingCognitive neuroscience of visual object recognitionColor imageSegmentation

摘要: This paper presents the vision system of robot soccer team Agilo RoboCuppers - RoboCup image understanding group (FG BV) at Technische Universitat Munchen.We present a fast and robust color classification method yielding significant regions in image. The boundaries between adjacent are used to localize objects like ball or other robots on field. Furthermore for each player free motion space is determined its position orientation field estimated. All this done completely based, without any additional sensors.

参考文章(4)
Michael Klupsch, Object-Oriented Representation of Time-Varying Data Sequences in Multiagent Systems International Institute of Informatics and Systemics (IIIS). ,(1998)
Thorsten Bandlow, Robert Hanek, Michael Klupsch, Thorsten Schmitt, Agilo RoboCuppers: RoboCup Team Description robot soccer world cup. pp. 567- 570 ,(2000) , 10.1007/3-540-45327-X_83
R.K. Lenz, R.Y. Tsai, Techniques for calibration of the scale factor and image center for high accuracy 3-D machine vision metrology IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 10, pp. 713- 720 ,(1988) , 10.1109/34.6781
Carsten T. Steger, Architecture for computer vision application development within the HORUS system Journal of Electronic Imaging. ,vol. 6, pp. 244- 261 ,(1997) , 10.1117/12.266862