Abstract
This paper presents a benchmark for evaluating the raster to vector conversion systems. The benchmark is designed for evaluating the performance of graphics recognition systems on images that contain polygons (solid) within the images. Our contribution is two-fold, an object mapping algorithm to spatially locate errors within the drawing and then a cycle graph matching distance that indicates the accuracy of the polygonal approximation. The performance incorporates many aspects and factors based on uniform units while the method remains non-rigid (thresholdless). This benchmark gives a scientific comparison at polygon level of coherency and uses practical performance evaluation methods that can be applied to complete polygonization systems. A system dedicated to cadastral map vectorization was evaluated under this benchmark and its performance results are presented in this paper. By stress testing a given system, we demonstrate that our protocol can reveal strengths and weaknesses of a system. The behavior of our set of indices was analyzed when increasing image degradation. We hope that this benchmark will help assessing the state of the art in graphics recognition and current vectorization technologies.
Similar content being viewed by others
References
Byrnes D.: Raster-to-vector comes of age with auto CAD release. Cadalyst 14, 48–70 (1997)
Kasturi, R., Tombre, K. (eds.): Graphics recognition: methods and applications. In: First International Workshop, University Park, PA, USA, August 1995. Selected Papers Published as Lecture Notes in Computer Science, p. 1072 (1996)
Kong B., Phillips I.T., Haralick R.M., Prasad A., Kasturi R.: A benchmark: performance evaluation of dashed-line detection algorithms. Graphics recognition—methods and applications. Lect. Notes Comput. Sci. 1072, 270–285 (1996)
Dori, D., Wenyin, L., Peleg, M.: How to win a dashed line detection contest. Graphics Recognition Methods and Applications. Lecture Notes in Computer Science, Springer, p. 1072 (1996)
Chhabra, A., Phillips, I.: The Second International Graphics Recognition Contest—Raster to Vector Conversion: A Report. Graphics Recognition: Algorithms and Systems. Lecture Notes in Computer Science, Springer, p. 1389 (1998)
Phillips, I., Liang, J., Chhabra, A., Haralick, R.: A Performance Evaluation Protocol for Graphics Recognition Systems. Graphics Recognition: Algorithms and Systems. Lecture Notes in Computer Science, Springer, p. 1389 (1998)
Phillips I., Chhabra A.: Empirical performance evaluation of graphics recognition systems. IEEE Trans. Pattern Anal. Mach. Intell. 21(9), 849–870 (1999)
Wenyin L., Dori D.: A protocol for performance evaluation of line detection algorithms. Mach. Vis. Appl. 9(5), 240–250 (1997)
Haralick R.M.: Performance characterization in image analysis thinning, a case in point. Pattern Recognit. Lett. 13, 5–12 (1992)
Lee, S., Lam, L., Suen, C.Y.: Performance evaluation of skeletonization algorithms for document image processing. In: Proceedings of the First International Conference on Document Analysis and Recognition, vol. 1, pp. 260–271 (1991)
Lam, L., Suen, C.Y.: Evaluation of thinning algorithms from an OCR viewpoint. In: Proceedings of the Second International Conference on Document Analysis and Recognition vol. 1, pp. 287–290 (1993)
Jaisimha, M.Y., Haralick, R.M., Dori, D.: A methodology for the characterization of the performance of thinning algorithms. In: Proceedings of the Second International Conference on Document Analysis and Recognition, vol. 1, pp. 282–286 (1993)
Cordella L.P., Marcelli A.: An alternative approach to the performance evaluation of thinning algorithms for document processing applications. Graphics recognition methods and applications. Lect. Notes Comput. Sci. 1072, 13–22 (1996)
Kasturi R., Bow S.T., El-Masri W., Shah J., Gattiker J.R., Mokate U.B.: A system for interpretation of line drawings. IEEE Trans. Pattern Anal Mach. Intell. 17, 978–992 (1990)
Nagasamy V., Langrana N.: Engineering drawing processing and vectorization system. Comput. Vis. Graph. Image Process 49, 379–397 (1990)
Filipski A.J., Flandrena R.: Automated conversion of engineering drawings to CAD form. Proc. IEEE 80, 1195–1209 (1992)
Boatto L.: An interpretation system for land register maps. IEEE Comput. 25(7), 25–32 (1992)
Vaxiviere P., Tombre K.: Celestin: CAD conversion of mechanical drawings. IEEE Comput. 25, 46–54 (1992)
Dori D.: Vector-based arc segmentation in the machine drawing understanding system environment. IEEE Trans. Pattern Anal Mach. Intell. 17, 959–971 (1995)
Dori D., Liang Y., Dowell J., Chai I.: Spare pixel recognition of primitives in engineering drawings. Mach. Vis. Appl. 6, 79–82 (1993)
Kasturi, R., Tombre K., (eds.): Graphics Recognition, Methods and Applications, First International Workshop, University Park, PA, USA, 10–11, August 1995. Selected Papers, volume 1072 of Lecture Notes in Computer Science. Springer (1996)
Graphics Recognition, Algorithms and Systems, Second International Workshop, GREC’97, Nancy, France, 22–23, August 1997, Selected Papers. In: Karl, T., Atul, K.C., (eds.) GREC, volume 1389 of Lecture Notes in Computer Science. Springer (1998)
Atul, K.C., Dori, D., (eds.): Graphics Recognition, Recent Advances, Third International Workshop, GREC’99 Jaipur, India, 26-27, September 1999, Selected Papers, volume 1941 of Lecture Notes in Computer Science. Springer (2000)
Hori O., Doermann D.S.: Quantitative measurement of the performance of raster-to-vector conversion algorithms. Graphics recognition—methods and applications. Lect. Notes Comput. Sci. 1072, 57–68 (1996)
Robert W.F.: Nondeterministic algorithms. J. ACM 14(4), 636–644 (1967)
Bunke H.: On a relation between graph edit distance and maximum common subgraph. Pattern Recogni. Lett. 18(9), 689–694 (1997)
Bunke H., Shearer K.: A graph distance metric based on the maximal common subgraph. Pattern Recogni. Lett. 19(3-4), 255–259 (1998)
Arkin E.M., Chew L.P., Huttenlocher D.P., Kedem K., Mitchell J.S.B.: An efficiently computable metric for comparing polygonal shapes. IEEE Trans. Pattern Anal. Mach. Intell. 13(3), 209–216 (1991)
Lladós J., Marti E., Villanueva J.J.: Symbol recognition by error-tolerant subgraph matching between region adjacency graphs. IEEE Trans. Pattern Anal. Mach. Intell. 23, 1137–1143 (2001)
Bunke H., Allermann G.: Inexact graph matching for structural pattern recognition. Pattern Recogni. Lett. 1, 245–253 (1983)
Sanfeliu A., Fu K.: A distance measure between attributed relational graphs for pattern recognition. IEEE Trans. Syst. Man Cybern. B Cybern. 1, 353–363 (1983)
Tsay Y.T., Tsai W.H.: Model-guided attributed string matching by split-and-merge for shape recognition. Int. J. Pattern Recogni. Art. Intell. 3(2), 159–179 (1989)
Valveny, E., Dosch, P.: Symbol Recognition Contest: a Synthesis. Graphics Recognition, pp. 368–385 (2004)
Valveny E., Dosch P., Winstanley A., Zhou Y., Yang Su., Yan L., Wenyin L., Elliman D., Delalandre M., Trupin E., Adam S., Ogier J.-M.: A general framework for the evaluation of symbol recognition methods. Int. J. Doc. Anal. Recogni. 9(1), 59–74 (2007)
Mathieu, D., Ernest, V., Tony, P., Dimosthenis, K.: Generation of synthetic documents for performance evaluation of symbol recognition and spotting systems. Int. J. Doc. Anal. Recogni. Page online first (2010)
Philippe, D., Ernest, V.: Report on the Second Symbol Recognition Contest (2006)
Cootes T.F., Taylor C.J., Cooper D.H., Graham J.: Active shape models—their training and application. Comput. Vis. Image Underst. 61(1), 38–59 (1995)
Ghosh D., Shivaprasad A.P.: An analytic approach for generation of artificial hand-printed character database from given generative models. Pattern recogni. 32(6), 907–920 (1999)
Valveny E., Marti E.: A model for image generation and symbol recognition through the deformation of lineal shapes. Pattern Recogni. Lett. 24(15), 2857–2867 (2003)
Robert, M.H., Henry, S.B., Adihan, D.: Document Degradation Models: Parameter Estimation and Model Validation (2009)
Sanniti di Baja G., Thiel E.: Skeletonization algorithm running on path-based distance maps. Image Vis. Comput. 14(1), 47–57 (1996)
Wall K., Danielsson P.-E.: A fast sequential method for polygonal approximation of digitized curves. Comput. Vision Graph. Image Process 28(3), 220–227 (1984)
Romain R., Jean-Christophe, B., Jean-Marc, O.: A colour text/graphics separation based on a graph representation. In: 19th International Conference on Pattern Recognition (ICPR), pp. 1–4 (2008)
Romain R., Jean-Christophe, B., Jean-Marc, O.: Object Extraction from Colour Cadastral Maps. In: IAPR International Workshop on Document Analysis Systems, pp. 506–514 (2008)
Hervé, L., Romain, R., Sébastien, A., Yves, L., Pierre, H., Éric, T.: Approximation of Digital Curves using a Multi-Objective Genetic Algorithm. In: 18th International Conference on Pattern Recognition (ICPR), pp. 716–719, Washington, DC, USA, (2006). IEEE Computer Society
Ferreira, A. Jr, Manuel Jr., Fonseca, J., Jorge, J.A.: Polygon Detection from a Set of Lines. In: Proceedings of 12 o Encontro Português de Computação Gráfica (12th EPCG), pages 159–162 (2003)
Raveaux, R., Burie, J.-C., Ogier, J.-M.: A Colour Document Interpretation: Application to Ancient Cadastral Maps. International Conference on Document Analysis and Recognition, vol. 2, pp. 1128–1132 (2007)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Raveaux, R., Burie, J.C. & Ogier, J.M. A local evaluation of vectorized documents by means of polygon assignments and matching. IJDAR 15, 21–43 (2012). https://doi.org/10.1007/s10032-010-0143-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10032-010-0143-3