Abstract
Visual recognition of microelectromechanical parts is necessary for automation of the assembly process. The visual recognition system that we have developed is based partly on neural networks and partly on digital image-processing techniques. The system takes grey-level microscope images and produces recognition code as the output as well as information about micropart position. The recognition procedure is not sensitive to micropart position. This is ensured by preprocessing based on calculation of the image moment properties. For the recognition, a supervised feedforward neural network is utilized. A combination of standard backpropagation and resilient propagation is chosen for learning the network. The performance of the system is tested on recognition of the parts of a microvalve system. The results are satisfactory with respect to recognition accuracy and recognition time.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Becker, E. W., Ehrfeld, W., Hagmann, P., Maner, A. and Münchmeyer, D. (1986) Fabrication of microstructures with high aspect ratios and great structural heights by synchrotron radiation lithography. Microelectronics Engineering, 4, 35–56.
Braun, H. and Riedmiller, M. (1992) RPROP: a fast adaptive learning algorithm, in Proceedings of the International Symposium on Computer and Information Science VII.
Bryzek, J., Petersen, K. and McCulley, W. (1994) Micromachines on the march. IEEE Spectrum, 5, 20–32.
Cios, K. J. and Shin, I. (1995) Image recognition neural network: IRNN. Neurocomputing, 7, 159–185.
Fahrenberg, J., Maas, D., Menz, W. and Schomburg, W. K. (1994) Active microvalve system manufactured by the LIGA process, in Proceedings of ACTUATOR `94, Bremen, Germany, pp. 71–74.
Gengenbach, U. and Göttert, J. (1994) Requirements and first results of an assembly and handling apparatus for automated fabrication of microsystems, in Proceedings of Seminar on handling and Assembly of Microparts, Vienna, Austria, IFWT, TU Wien.
Gonzales, R. and Woods, R. (1992) Digital Image Processing, Addison-Wesley, Reading, MA.
Hanseler, J. (1995) Back propagation, in Artificial Neural Networks, Braspenning, P. J., Thuijsman, F. and Weijters, A.J.M.M. (eds), Springer, pp. 37–66.
Kim, H. and Nam., K. (1995) Object recognition of one-DOF tools by a back-propagation neural net. IEEE Transactions on Neural Networks, 2, 484–487.
Kolesar, Jr, E. S. and Dyson, C. S. (1995) Optical imaging with a piezoelectric robotic tactile sensor. IEEE Journal of Micro-electromechanical Systems, 2, 87–96.
Maas, D., Büstgens, B., Fahrenberg, J., Keller, W. and Seidel, D. (1994) Application of adhesive bonding for integration of microfluidic components, in Proceedings ACTUATOR `94, Bremen, Germany, pp. 75–78.
Menz, W., Bacher, W., Harmening, M. and Michel, A. (1991) The LIGA technique–a novel concept for microstructures and the combination with Si-technologies by injection molding, in Proceedings of MEMS 91, Nara, Japan, pp. 69–73.
Mitsuishi, M., Nagao, T., Natamura, Y., Kramer, B. and Warisawa, S. (1992) A manufacturing system for the global age, in Human Aspects and Computer Integrated Manufacturing Olling, G. J. and Kimura, F. (eds.), Elsevier Science, pp. 841–852.
Musavi, M. T., Chan, K. H., Hummels, D. M. and Kalantri, K. (1994) On the generalization ability of neural network classifiers. IEEE Transactions on Pattern Recognition and Machine Intelligence, 6, 659–663.
Radjenović-Mrčarica, J. (1996) Neural network visual recognition system for microelectromechanical parts assembly, PhD Thesis, UBTUW-Verlag, Vienna, Austria.
Rosenfeld, A. and Kak, A. (1989) Digital Image Processing.
Rumelhart, D. E. Hinton, G. E. and Williams, R. J. (1986) Learning internal representation by error propagation, in Parallel Distributed Processing: Explorations in the Micro-structures Cognition, Rumelhart, D. E. and Mc.Clelland, J. L. (eds.) MIT Press, Cambrige, MA, pp. 318–362.
Seithi, I. K. and Jain A. K. (1991) Artificial Neural Networks and Statistical Pattern Recognition–Old and New Connections, Elsevier Science.
Wilson, S. S. (1991) Teaching network conectivity using simulated annealing on a massively parallel processor, Proceedings of the IEEE, 4, 559–566.
Zell, A., Mamier, G., Vogt, M., Mache, N., Hübner, R., Herrmann, K. U., Soyez, T., Schmalzl, M., Sommer, T., Hatzigeorgiou, A., Dörning, S. and Posselt, D. (1995) SNNS: Stuttgart Neural Network Simulator: Users Manual, Version 4.0, University of Stuttgart, Institute of Applied and Distributed High Performance Systems.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
RADJENOVIC´-MRČARICA , J., DETTER , H. Recognition of thin, flat microelectromechanical structures for automation of the assembly process. Journal of Intelligent Manufacturing 8, 191–201 (1997). https://doi.org/10.1023/A:1018569123830
Issue Date:
DOI: https://doi.org/10.1023/A:1018569123830