GREC’09 Arc Segmentation Contest

GREC'09 organized by IAPR will be held at City University of La Rochelle, France. As in previous GREC events, an Arc Segmentation Contest is also organized (see previous contest web page). GREC'09 Arc Segmentation Contest will be organized by School of Computer Sciences / Universiti Sains Malaysia, Malaysia. Therefore, we would like to invite you to participate in this contest by providing the output of raster to vector methods. Both research prototypes and commercial software are accepted. The output of the method could be provided to us in VEC format. However, files in DXF format are also accepted. We also encourage researchers to provide us with any theme/idea to enrich this contest. Some suggested themes (not limited to) are as follows:

  1. The use of noise with different levels.
  2. The use of more specific performance evaluation. e.g. studying vector detection and false alarm of contested methods separately.
  3. Testing of several research prototypes (provide name if possible).
  4. The use of more commercial software (provide name if possible).
  5. Within-method comparison, i.e. studying the performance of different versions of the same method (developed over time) to gauge the improvement in the performance of line detection.
  6. Recommendation of a performance evaluation method: VRI[1], Vectorial Score[2], or other?
  7. Any other suggestion?

You can e-mail your suggestions or results to Abdullah Zawawi or to Hasan.

GREC'09 Arc Segmentation Contest- A Proposition

Training Images :
Five scanned drawings in TIF format (LZW compression) with their corresponding ground truth images in VEC format are provided in this ZIP file.

Test Images :
Seven scanned drawings in TIF format (LZW compression) with their corresponding ground truth images in VEC format are provided in this ZIP file.


References :  
[1] W. Y. Liu and D. Dori, "A protocol for performance evaluation of line detection algorithms," Machine Vision and Applications, vol. 9, pp. 240-250, 1997.
[2] F. Shafait, D. Keysers, and T. M. Breuel, "Pixel-accurate representation and evaluation of page segmentation in document images," 18th International Conference on Pattern Recognition, Vol 1, Proceedings, pp. 872-875, 2006.