A computer vision system for automated corn seed purity analysis

ABSTRACT

Electrophoresis gel analysis is a viable technique for determining the purity of hybrid corn seeds. Visually analyzing the electrophoretic gel images is a very tedious and time-consuming task. In this paper, a computer vision system integrating image processing and pattern recognition techniques with domain-specific structural information to automate the electrophoresis gel scoring procedure is presented. A set of image processing algorithms are developed to perform extraction of the region of interest, segmentation of samples, identification of bands within samples, and final classification of different types of seeds. The image processing algorithms utilize the structural information and operator expertise to achieve high classification rate with fuzzy and incomplete information contained on the electrophoresis gels. The developed technique clearly demonstrates the potential of using computer vision in automating the gel scoring procedure. The developed technology may also be extended to other areas such as general one-dimensional electrophoresis and high performance thin layer chromatography.

REFERENCES

• 1: Agard, D. A., Steinberg, R. A., and S t roud, R. M., "Quantitative Analysis of Electrophoretograms: A Mathematical Approach to Super-Resolution," Analytical Biochemistry III, 1981.
• 2: Anderson, N. L., Taylor, J., Scandora, A. E., Coulter, B. P., and Anderson, N. G., "The TY(}HO System for Computer Analysis of TWo-Dimensional Gel Electrophoresis Patterns," Clinical Chemistry, Vol 27, No. 11, 1981.
• 3: Brown, W. T., and Ezer, A., "A Cc~ute r Program U~ing Gaussian Fitting for EValuation of Two-Dimensional Gels," Clinical Chemistry, Vol. 28, No. 4, 1982.
• 4: Cardy, B. J., $tuber, C. W., and Goodman, M. M., Techniques for Starch Gel Electrophoresis -~ EnzYmes fr-om Maize. Memo No. 131T, Institute o~ statistics, North Carolina State University.
• 5: Garrels, J. I.,"TWo-Dimensional Gel Electrophoresis and Computer Analysis of Proteins Synthesized by Clonal Cell Lines," The Journal of Biological Chemistry, Vol. 254, No. 16, 1979.
• 6: Goodman, M. M., S tuber, C. W., "Genetic Identification of Lines and Crosses Using Isoenzlm~ Electrophoresis," Proc American Seed Trade Association Corn and Sorghum Conference, 1980.
• 7: Lemkin, P. F., and Lipkin, L. E., "GELLAB- A Computer System for 2-D Gel Electrophoresis Analysis I. Segmen~ti~ of Spots and System Preliminaries," Compute rs and Biomedical Research, VoI. 14, 1981-}-.
• 8: Lin, W. C., Chung, J., Litt, M., Automated Electrophoresis Gel Scoring: A Preliminary Study. Topical Report No.7199~ The Standard Oil Company, R&D. June 5, 1985.
• 9: Skolnick, M. M., Sternberg, S. R., and Neel, J. V., "Computer Programs for Adapting Two-Dimensional Gels to the Study of Mutation," Clinical Chemistry, Vol. 28, No. 4, 1982.
• 10: Smith, S., "The Testing of Seed Lot Purity by Electrophoresis and Other Applications to Maize Breeding," Maize Genetics Coope ration Newsletter, Vol. 10'7, 1980-1981.
• 11: Sternberg, S. R., "Biomedical Image Processing,"IEEE C0mputer Januaryv 1983.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.5 PATTERN RECOGNITION
      I.5.4 Applications
          Subjects: Computer vision

Additional Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.6 Segmentation
          Subjects: Region growing, partitioning
  I.5 PATTERN RECOGNITION
      I.5.2 Design Methodology
          Subjects: Pattern analysis

  J. Computer Applications
  J.2 PHYSICAL SCIENCES AND ENGINEERING
      Subjects: Chemistry

General Terms:
Algorithms