Diploma Thesis DIP-3016

BibliographyBonilla, Daniel Rubio: Automatic color segmentation enhancement based on world knowledge.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Diploma Thesis No. 3016 (2010).
73 pages, english.
CR-SchemaC.1.4 (Processor Architectures, Parallel Architectures)
I.2.9 (Robotics)
I.2.10 (Vision and Scene Understanding)
I.4.8 (Image Processing and Computer Vision Scene Analysis)
I.5.3 (Pattern Recognition Clustering)
Abstract

This project started with the aim of developing a faster, more precise and automatic system for calibrating the colors of the football field (the environment in which the robots, the “football players”, that participate in the RoboCup have to operate), so that the images taken from the camera (what the robots “see”) are segmented and can be later processed by another “world understanding” software. After a first evaluation of the different alternatives we had to develop this project for us to the wanted results, we decided that our best option was to use a clustering approach. It consists in grouping the colors in different clusters, and then select the clusters with more possibility to be a specific color representing each object of the real world.

Full text and
other links
PDF (4042836 Bytes)
Access to students' publications restricted to the faculty due to current privacy regulations
Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Parallel Systems
Superviser(s)Koch, Andreas
Entry dateApril 27, 2010
   Publ. Computer Science