A versatile visual inspection tool for the manufacturing process

The dynamically changing nature and the complex behaviour of processes in manufacturing cells dictate the need for lean, agile and flexible manufacturing systems. This research describes a versatile, intelligent vision system capable of performing a variety of tasks for the manufacturing process. The key features of the system are reconfigurability, adaptation, and real-time performance. It is based on higher-order neural networks (HONNs), whose structure is designed using a priori information related to the expected relationships between input pixels. The incorporation of prior information is the reason that HONNs demonstrate invariance to certain geometric distortions. An input representation scheme known as coarse coding was used to represent the fine resolution image with a set of offset overlaying coarse resolution images. The performance of the system is examined in three real-world application areas, i.e. the classification of screws, the classification of rivets, and finally the detection of flaws in axisymmetric engineering parts.<<ETX>>

[1]  Bernard Widrow,et al.  MADALINE RULE II: a training algorithm for neural networks , 1988, ICNN.

[2]  Long Quan,et al.  Projective Invariants for Vision , 1992 .

[3]  Georgios B. Giannakis,et al.  Object and Texture Classification Using Higher Order Statistics , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  M. B. Zarrop,et al.  Higher-order neural networks for invariant blemish detection , 1993 .

[5]  Georgios B. Giannakis,et al.  Signal detection and classification using matched filtering and higher order statistics , 1989, IEEE Trans. Acoust. Speech Signal Process..

[6]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[7]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[8]  Etienne Barnard,et al.  Invariance and neural nets , 1991, IEEE Trans. Neural Networks.

[9]  Colin Giles,et al.  Learning, invariance, and generalization in high-order neural networks. , 1987, Applied optics.

[10]  S. K. Rogers,et al.  The use of neural networks in PSRI target recognition , 1988, IEEE 1988 International Conference on Neural Networks.

[11]  Kunihiko Fukushima,et al.  Analysis of the process of visual pattern recognition by the neocognitron , 1989, Neural Networks.

[12]  C. Lee Giles,et al.  Nonlinear dynamics of artificial neural systems , 1987 .

[13]  P. E. Wellstead,et al.  Visual inspection of axisymmetric parts , 1993 .

[14]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .