Artificial Intelligence as an Approach to Improve Ultrasonic Log Scanning

  • Wei Han
  • Rolf Birkeland
Part of the Acoustical Imaging book series (ACIM, volume 20)

Abstract

Scanning of logs for optimum material utilization becomes more and more important in wood industry. Ultrasound tomography seems to be a good solution to log scanning because of its capability of scrutinizing a log both externally and internally or the external geometry and the internal defects. However, the substantial computations and the complex tomogram translation make the industrial application difficult. This paper presents principles of applying artificial intelligence (AI) to ultrasonic scanning for the purpose of improving scanning efficiency. The presentation involves the integration of AI in signal acquisition and treatment, and defect computation. The AI functions involved include learning, knowledge-based reasoning, fuzziness handling and neural network application. An intelligent ultrasonic system is introduced to illustrate the principles discussed. The results given by the illustrating system indicate that the combination of ultrasound and AI opens up a promising future for a practical log scanning solution.

Keywords

Ultrasound AI log scanning 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Caudill, M. & C. Butler. 1990. Naturally Intelligent Systems. A Bradford Book, the MIT Press.Google Scholar
  2. 2.
    Han, W. 1991. An intelligent ultrasonic system for log scanning. Ph. D. dissertation submitted to the Norwegian Institute of Technology, Trondheim, Norway.Google Scholar
  3. 3.
    Han W. & R. Birkeland. 1992. Ultrasonic scanning of logs. Industrial Metrology 2:253–281.CrossRefGoogle Scholar
  4. 4.
    Yoshiaki, Y. & J. I. Tsujii. 1982. Artificial Intelligence: Concepts, Techniques and Applications. John Wiley & Sons, Chichester.Google Scholar
  5. 5.
    Zadah, L. A. 1965. Fuzzy sets. Information and Control 8:338–353.MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1993

Authors and Affiliations

  • Wei Han
    • 1
  • Rolf Birkeland
    • 1
  1. 1.The Norwegian Institute of Wood TechnologyOslo 3Norway

Personalised recommendations