Automatic machine vision calibration using statistical and neural network methods

A methodology is presented for camera calibration that is designed to improve the accuracy of machine vision based object measurement systems. The regression and artificial neural network techniques studied are considered to be complimentary rather than competitive. Neural networks have been identified as being particularly useful for the precise modelling of non-linear response, and offer the additional benefits of being non-prescriptive and generally applicable to factors such as radial lens distortion, manufacturing errors and minor camera misalignments. The combination of these modelling techniques within automated program control strategy is suggested as a new approach for straightforward and accessible machine vision calibration. The method has particularly good application to vision metrology and reverse engineering tasks. A demonstrator system has been constructed, employing a scanning laser line and vision system for object measurement in three-dimensions. Experimental results are presented along with a demonstration of the reduction in measurement error that can be attained through the application of regression analysis and artificial neural network modelling.

[1]  Kenneth J.A. Brookes.  Sintered hard materials on show at MACH 2000 , 2000 .

[2]  Martin T. Hagan,et al.  Neural network design , 1995 .

[3]  Seong-Whan Lee,et al.  Biologically Motivated Computer Vision , 2002, Lecture Notes in Computer Science.

[4]  Randall M. German,et al.  Experimental analysis and neural network modelling of the rheological behaviour of powder injection moulding feedstocks formed with bimodal powder mixtures , 2000 .

[5]  Sheng-Wen Shih,et al.  When Should We Consider Lens Distortion in Camera Calibration , 1995, MVA.

[6]  Reinhard Klette,et al.  Computer vision - three-dimensional data from images , 1998 .

[7]  Ernest L. Hall,et al.  Techniques for fisheye lens calibration using a minimal number of measurements , 2000, SPIE Optics East.

[8]  Lakhmi C. Jain,et al.  Industrial Applications of Neural Networks , 1998 .

[9]  J. Alison Noble,et al.  From inspection to process understanding and monitoring: a view on computer vision in manufacturing , 1995, Image Vis. Comput..

[10]  David W. Coit,et al.  STATIC NEURAL NETWORK PROCESS MODELS : CONSIDERATIONS AND CASE STUDIES , 1998 .

[11]  A. Matthews,et al.  Taguchi and TQM: Quality issues for surface engineered applications , 1998 .

[12]  Madan M. Gupta,et al.  Neuro-Vision Systems: Principles and Applications , 1995 .

[13]  Roger Y. Tsai,et al.  A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses , 1987, IEEE J. Robotics Autom..

[14]  Lyndon N. Smith,et al.  Experimental Study and Neural Network Modeling of the Stability of Powder Injection Molding Feedstocks , 2000 .

[15]  Vittorio Murino,et al.  Artificial Neural Networks for Image Analysis and Computer Vision , 2001, Image Vis. Comput..

[16]  Jerry Y. H. Fuh,et al.  A neural network approach for early cost estimation of packaging products , 1998 .

[17]  László Monostori,et al.  Monitoring of the Milling Process through Neural Network and Fuzzy Techniques , 1993, KNOWHSEM.

[18]  Barry Lennox,et al.  Industrial application of neural networks — an investigation , 2001 .

[19]  V. H. Chan,et al.  A multi-sensor approach to automating co-ordinate measuring machine-based reverse engineering , 2001 .

[20]  Wei-Song Lin,et al.  Accurate linear technique for camera calibration considering lens distortion by solving an eigenvalue problem , 1993 .

[21]  Heinrich Niemann,et al.  Neural networks for the recognition and pose estimation of 3D objects from a single 2D perspective view , 2001, Image Vis. Comput..

[22]  Melvyn L. Smith,et al.  Gradient space analysis of surface defects using a photometric stereo derived bump map , 1999, Image Vis. Comput..

[23]  Melvyn L. Smith,et al.  The analysis of surface texture using photometric stereo acquisition and gradient space domain mapping , 1999, Image Vis. Comput..

[24]  Azriel Rosenfeld,et al.  Computer Vision , 1988, Adv. Comput..

[25]  Engelbert Westkämper,et al.  Computer-assisted manufacturing process optimization with neural networks , 1998, J. Intell. Manuf..

[26]  Chao-Ton Su,et al.  A comparison of statistical regression and neural network methods in modeling measurement errors for computer vision inspection systems , 1995 .

[27]  Christian Wöhler,et al.  Real-time object recognition on image sequences with the adaptable time delay neural network algorithm - applications for autonomous vehicles , 2001, Image Vis. Comput..

[28]  Ming-Chuan Leu,et al.  Design of optical triangulation devices , 1989 .

[29]  Richard Stamp,et al.  Automated inspection of textured ceramic tiles , 2000 .

[30]  M. A. Younis.  On line surface roughness measurements using image processing towards an adaptive control , 1998 .

[31]  Gideon Cohen,et al.  Neural networks implementations to control real-time manufacturing systems , 1998 .

[32]  V. Kvasnicka,et al.  Neural and Adaptive Systems: Fundamentals Through Simulations , 2001, IEEE Trans. Neural Networks.

[33]  Godwin J. Udo,et al.  Neural networks applications in manufacturing processes , 1992 .

[34]  Jose C. Principe,et al.  Neural and Adaptive Systems: Fundamentals through Simulations with CD-ROM , 1999 .

[35]  S. H. Huang,et al.  Applications of neural networks in manufacturing: a state-of-the-art survey , 1995 .

[36]  Ryan G. Rosandich.  Intelligent visual inspection : using artificial neural networks , 1997 .

[37]  Chao-Ton Su,et al.  Neural networks for precise measurement in computer vision systems , 1995 .

[38]  Tim J. Gallwey.  Evaluation and control of industrial inspection: Part I – Guidelines for the practitioner , 1998 .

[39]  Richard J. Cant,et al.  A framework for the evaluation of volume rendering techniques on a task specific basis using neural networks , 2001, Comput. Graph..

[40]  J. D. T. Tannock,et al.  A review of neural networks for statistical process control , 1998, J. Intell. Manuf..

[41]  Melvyn L. Smith,et al.  Surface texture analysis based upon the visually acquired perturbation of surface normals , 1997, Image Vis. Comput..

[42]  Cihan H. Dagli,et al.  The use of feedforward neural networks for machine vision calibration , 1999 .

[43]  Amit Singhal,et al.  Computer Vision and Fuzzy-Neural Systems , 2004, J. Electronic Imaging.

[44]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[45]  Claus P. Keferstein,et al.  Testing bench for laser triangulation sensors , 1998 .