Adaptive Algorithms Based on Technical Vision - Neural Networks and Micromechanics - page 215

could be explained by the fact that each object in the image contains many pixels
and these pixels give much more detailed information than one pixel.
A computer vision system for improving the microassembly device's precision
is proposed. Two algorithms of image recognition (neural classifier and neural
interpolator) were tested in the task of pin-hole relative position detection. The
neural classifier permits us to recognize the displacement of the pin relative to the
hole with 1 pixel tolerance, while the neural interpolator permits recognition with
0.5 pixel tolerance. The absolute values of detectable displacements depend on the
optical channel resolution. In our case, one pixel corresponds to 0.05 mm X -axis
displacements and 0.1 mm Y -axis displacements. This precision is sufficient for
many different assembly processes.
References
1. Kussul E., Baidyk T., Ruiz-Huerta L., Caballero A., Velasco G., Kasatkina L., 2002, Devel-
opment of Micromachine Tool Prototypes for Microfactories, Journal of Micromechanics and
Microengineering , 12, pp. 795-813.
2. Kussul E., Baidyk T. Neural Random Threshold Classifier in OCR Application. In: Proceedings
of the Second All-Ukrainian International Conference, Kiev, Ukraine, 1994, pp. 154-157.
3. Kussul E., Baidyk T., Lukovitch V., Rachkovskij D. Adaptive High Performance Classifier
Based on Random Threshold Neurons. In R. Trappl (ed.) Cybernetics and Systems'94,
Singapore: World Scientific Publishing Co. Pte. Ltd., 1994, pp. 1687-1695.
4. Baidyk T. Application of Flat Image Recognition Technique for Automation of Micro Device
Production. In: Proceedings of 2001 IEEE/ASME International Conference on Advanced
Intelligent Mechatronics, Teatro Sociale Como, Italy, 2001, pp. 488-494.
5. Baidyk T., Kussul E. Application of Neural Classifier for Flat Image Recognition in the
Process of Microdevice Assembly. In: Proceedings of IEEE International Joint Conference on
Neural Networks, Hawaii, USA, Vol. 1, 2002, pp. 160-164.
6. Baidyk T., Kussul E., Makeyev O., Caballero A., Ruiz L., Carrera G., Velasco G. Flat Image
Recognition in the Process of Microdevice Assembly. Pattern Recognition Letters , Vol. 25/1,
2004, pp. 107-118.
7. Bottou L., Cortes C., Denker J., Drucker H., Guyon L., Jackel L., LeCun Y., Muller U.,
Sackinger E., Simard P., Vapnik V., Comparison of Classifier Methods: A Case Study in
Handwritten Digit Recognition, Proc. of 12th IAPR Intern. Conf. on Pattern Recognition,
1994, Vol. 2, pp. 77-82.
8. Makeyev O., Neural Interpolator for Image Recognition in the Process of Microdevice
Assembly. In: IEEE IJCNN'2003, Portland, Oregon, USA, July 20-24, Vol. 3, 2003, pp.
2222-2226.
9. Baidyk T., Kussul E., Makeyev O. Image Recognition System for Microdevice Assembly,
Twenty-First IASTED International Multi-Conference on Applied Informatics, AI2003,
Innsbruck, Austria, 2003, pp. 243-248.
10. Kussul E., Baidyk T., 2004, Improved Method of Handwritten Digit Recognition Tested on
MNIST Database. Image and Vision Computing , Vol. 22/12, pp. 971-981.
11. Kussul E., Baidyk T., 2006, LIRA Neural Classifier for Handwritten Digit Recognition
and Visual Controlled Microassembly. Neurocomputing , Vol. 69, Issue 16-18, 2006,
pp. 2227-2235.