Edge Detection of Petrographic Images Using Genetic Programming
Inhaltsverzeichnis
Reference
Brian J. Ross, F. Fueten, and D.Y. Yashkir: Edge Detection of Petrographic Images Using Genetic Programming. GECCO 2000, ed. D. Whitley et al., Morgan Kaufmann, 2000, pp. 658-665
DOI
Abstract
This paper discusses work in progress that uses genetic programming to evolve edge de- tectors for petrographic images. Microscopic images of thin sections from mineral samples are obtained using a rotating polarizer mi- croscope. These images are then processed using a number of lters, resulting in a set of nine ltered image parameters. In order to be useful for higher{level analysis, such as automatic mineral identi cation, the grain boundaries within these images must be iden- ti ed. Using genetic programming, edge de- tecting functions are evolved for this purpose. The edge detectors may use as any of the l- tered image parameters as input. Since the source images are large, a subset of the im- ages is sampled for training, and the remain- der of the image is used for testing. This training data is selected with a biased ran- dom sampling strategy. The complexity of the images dictates that a generic edge de- tector for all mineral specimens is infeasible. Rather, the most useful edge detectors will be those that are specialized for particular families of mineral specimens.
Extended Abstract
Bibtex
Used References
Canny, J. (1986). A computational approach to edge detection. IEEE Transactions on Pattern Analy- sis and Machine Learning 8(6), 679{698.
Daida, J.M., T. F. Nersano-Begey, S.J. Ross and J.F. Vesecky (1996). Computer-Assisted Design of Image Classi cation Algorithms: Dynamic and Static Fitness Evaluations in a Sca olded Ge- netic Programming Environment. In: Proc. Ge- netic Programming 1996 (J.R. Koza et al., Ed.). MIT Press. pp. 279{284.
Fueten, F. (1997). A computer controlled rotating polarizer stage for the petrographic microscope. Computers and Geosciences 23, 203{208.
Fueten, F. and S. Thompson (2000). Petrographic im- age edge detection using arti cial neural nets. In preparation.
Gonzalez, R.C. and R.E. Woods (1992). Digital Image Processing. Addison-Wesley.
Goodchild, J.S. and F. Fueten (1998). Edge detection in petrographic images using the rotating polar- izer stage. Computers and Geosciences 24, 745{ 751.
Harris, C. and B. Buxton (1996). Evolving Edge De- tectors with Genetic Programming. In: Proc. Ge- netic Programming 1996 (J.R. Koza et al., Ed.). MIT Press. pp. 309{314.
Koza, J.R. (1992). Genetic Programming. MIT Press. Lucier, B.J., S. Mamillapalli and J. Palsberg (1998). Program Optimization for Faster Genetic Pro- gramming. In: Proc. Genetic Programming 1998 (J.R. Koza et al, Ed.). Morgan Kaufmann. pp. 202{207.
Montana, D.J. (1995). Strongly Typed Genetic Pro- gramming. Evolutionary Computation 3(2), 199{ 230.
Poli, R. (1996). Genetic Programming for Image Analysis. In: Proc. Genetic Programming 1996 (J.R. Koza et al., Ed.). MIT Press. pp. 363{368.
Ross, B.J. (2000). The E ects of Randomly Sampled Training Data on Program Evolution. In: Proc. GECCO 2000 (D. Whitley et al., Ed.). Morgan Kaufmann.
Ross, B.J., F. Fueten and D.Y. Yashkir (1999). Auto- matic Mineral Identi cation Using Genetic Pro- gramming. Technical Report CS-99-04. Brock University, Dept. of Computer Science.
Thompson, S., F. Fueten and D. Bockus (1999). Min- eral identi cation using arti cial neural networks and the rotating polarizer stage. In: Proceedings Geovision 99: International Symposium on Imag- ing Applications in Geology. pp. 225{228.
Winkeler, J.F. and B.S. Manjunath (1997). Genetic Programming for Object Detection. In: Proc. Genetic Programming 1997 (J.R. Koza et al, Ed.). Morgan Kaufmann. Stanford University, CA, USA. pp. 330{335.
Zongker, D. and B. Punch (1995). lil-gp 1.0 User's Manual. Dept. of Computer Science, Michigan State University.
Links
Full Text
http://www.cosc.brocku.ca/~bross/research/rw047.ps
