MARCO: Active vision systems based in automatic learning for industrial applications
The objective of the project is the development of an active vision system with learning capabilities for industrial applications. The main issue to be resolved is the reduction of specific analysis, development and implementation of vision systems in, for example, unstructured environments (for auto-location applications in industrial or public buildings for automatic cleaning systems), or, for example, in environments where the access is extremely difficult (for maintenance tasks in the gas and water distribution industry). With this system we want to reduce the complexity of vision sensors in industrial applications and to extend its scope of use.
In the project relevant automatic learning techniques will be analysed to acquire the environment from a sequence of images obtained by means of a mobile robot, the minimum number of robust computer vision and pattern recognition techniques to be integrated in an active vision sensor will be studied, a prototype of the sensor will be developed and the use of the sensor in a pseudoindustrial work area will be verified.
The potential applications of such sensor are in the manufacturer, distribution and service industries. Moreover, the sensor is also applicable in the automotive industry.
A. Sanfeliu, F. Serratosa and R. Alquézar Mancho. Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition. International Journal of Pattern Recognition and Artificial Intelligence, 18(3): 375-396, 2004.
F. Serratosa, R. Alquézar Mancho and A. Sanfeliu. Function-described graphs for modelling objects represented by attributed graphs. Pattern Recognition, 36(3): 781-798, 2003.
A. Sanfeliu, R. Alquézar Mancho, J. Andrade-Cetto, J. Climent, F. Serratosa i Casanelles and J. Vergés. Graph-based representations and techniques for image processing and image analysis. Pattern Recognition, 35(3): 639-650, 2002.
J. Andrade-Cetto, T. Vidal-Calleja and A. Sanfeliu. Unscented transformation of vehicle states in SLAM, 2005 IEEE International Conference on Robotics and Automation, 2005, Barcelona, Espanya, pp. 323-328, IEEE.
J. Andrade-Cetto and A. Sanfeliu. The effects of partial observability in SLAM, 3a Jornada de la Red Temática RedVisión, 2004, Barcelona, Espanya, pp. 15-21, FME, UPC.
J. Andrade-Cetto and A. Sanfeliu. The effects of partial observability in SLAM, 2004 IEEE International Conference on Robotics and Automation, 2004, New Orleans, Estats Units d'Amèrica, pp. 397-402, IEEE.
J. Vergés-Llahí and A. Sanfeliu. A colour constancy algorithm based on the histogram of feasible colour mappings, 8th Iberoamerican Congress on Pattern Recognition, 2003, L'Havana, Cuba, in Progress in Pattern Recognition, Speech and Image Analysis, Vol 2905 of Lecture Notes in Computer Science, pp. 171-179, 2003, Springer, Berlin, Alemanya.
J. Andrade-Cetto and A. Sanfeliu. Learning of dynamic environments by a mobile robot from stereo cues, 2001 International Conference on Multisensor Fusion and Integration for Intelligent Systems, 2001, Baden-Baden, Germany, pp. 305-310, 2001, VDI/VDE-GMA, Duesseldorf, Alemanya.
J. Andrade-Cetto and A. Sanfeliu. Topological map learning for a mobile robot in indoor environments, 9th Spanish Symposium on Pattern Recognition and Image Analysis, 2001, Benicassim, Espanya, in Pattern recognition and image analysis: proceedings of the IX Spanish Symposium on Pattern Recognition and Image Analysis, Vol 6-7 of Treballs d'informàtica i tecnologia, pp. 221-226, 2001, UJI, Castelló de la Plana, Espanya.
A. Sanfeliu, J. Andrade-Cetto, R. Alquézar Mancho, J. Aranda, J. Climent, A. Grau, F. Serratosa and J. Vergés. MARCO: a mobile robot with learning capabilities to perceive and interact with its environment, 9th Spanish Symposium on Pattern Recognition and Image Analysis, 2001, Benicassim, Espanya, in Pattern recognition and image analysis: proceedings of the IX Spanish Symposium on Pattern Recognition and Image Analysis, Vol 6-7 of Treballs d'informàtica i tecnologia, pp. 219-224, 2001, UJI, Castelló de la Plana, Espanya.
F. Moreno-Noguer, J. Andrade-Cetto and A. Sanfeliu. Localization of human faces fusing color segmentation and depth from stereo, 8th IEEE International Conference on Emerging Technologies and Factory Automation, 2001, Antibes-Juan les Pins, França, pp. 527-535, IEEE.
J. Andrade-Cetto and A. Sanfeliu. Integration of perceptal grouping and depth, 15th International Conference on Pattern Recognition, 2000, Barcelona, Espanya, pp. 295-298, 2000, IEEE Computer Society, Barcelona, Espanya.
J. Vergés, J. Climent and A. Sanfeliu. Colour image segmentation solving hard-constraints on graph partitioning greedy algorithms, 15th International Conference on Pattern Recognition, 2000, Barcelona, Espanya, pp. 625-628, 2000, IEEE Computer Society, Barcelona, Espanya.
R. Alquézar Mancho, F. Serratosa and A. Sanfeliu. Distance between attributed graphs and function-described graphs relaxing 2nd order restrictions, 8th Joint IAPR International Workshop on Structural, Syntactic and Statistical Pattern Recognition, 2000, Alicante, Spain, in Advances in Pattern Recognition, Vol 1876 of Lecture Notes in Computer Science, pp. 277-286, 2000, Springer, Berlin.
A. Sanfeliu, A. Grau, J. Climent, R. Alquézar Mancho, F. Serratosa, J. Aranda, J. Vergés and J. Andrade-Cetto. Pattern recognition research at the IRI-CSIC/ESAII group, 2000 Foro Iberoamericano de Reconocimiento de Formas y Análisis de Imágenes, 2000, Barcelona, Spain, in Pattern recognition advances in Iberoamerica: foro 2000, pp. 345-354, 2000, UPC, Barcelona, Espanya.