Abstract
Keyword Spotting (KWS) improves the accessibility to handwritten historical documents by unconstrained retrievals of keywords. The proposed KWS framework operates on segmented words that are in turn represented as graphs. The actual KWS process is based on matching graphs by means of a cubic-time graph matching algorithm. Although this matching algorithm is quite efficient, the polynomial time complexity might still be a limiting factor (especially in case of large documents). The present paper introduces a novel approach that aims at speeding up the retrieval process. The basic idea is to first segment individual graphs into smaller subgraphs by means of a quadtree procedure. Eventually, the graph matching procedure can be conducted on the resulting pairs of smaller subgraphs. In an experimental evaluation on two benchmark datasets we empirically confirm substantial speed-ups while the KWS accuracy is nearly not affected.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
George Washington Papers at the Library of Congress, 1741–1799: Series 2, Letterbook 1, pp. 270–279 & pp. 300–309, http://memory.loc.gov/ammem/gwhtml/gwseries2.html.
- 2.
Parzival at IAM historical document database, http://www.fki.inf.unibe.ch/databases/iam-historical-document-database/parzival-database.
- 3.
BP stand for bipartite (LSAPs are also termed bipartite matching problem).
- 4.
We carry out our experiments on a high performance computing cluster with dozens of 2.2Â GHz CPU nodes. Hence, these readings refer to the average matching time per keyword measured in a sequential scenario.
References
Fernandez-Mota, D., Almazan, J., Cirera, N., Fornes, A., Llados, J.: BH2M: The Barcelona historical, handwritten marriages database. In: International Conference on Pattern Recognition, pp. 256–261 (2014)
Fischer, A., Frinken, V., Fornés, A., Bunke, H.: Transcription alignment of Latin manuscripts using hidden Markov models. In: Workshop on Historical Document Imaging and Processing, New York, p. 29 (2011)
Fischer, A., Keller, A., Frinken, V., Bunke, H.: Lexicon-free handwritten word spotting using character HMMs. Pattern Recognit. Lett. 33(7), 934–942 (2012)
Manmatha, R., Han, C., Riseman, E.: Word spotting: a new approach to indexing handwriting. In: Computer Vision and Pattern Recognition, pp. 631–637 (1996)
Rath, T., Manmatha, R.: Word image matching using dynamic time warping. In: Computer Vision and Pattern Recognition, vol. 2, pp. II-521–II-527 (2003)
RodrĂguez-Serrano, J.A., Perronnin, F.: Handwritten word-spotting using hidden Markov models and universal vocabularies. Pattern Recognit. 42(9), 2106–2116 (2009)
Rodriguez, J.A., Perronnin, F.: Local gradient histogram features for word spotting in unconstrained handwritten documents. In: International Conference on Frontiers in Handwriting Recognition, pp. 7–12 (2008)
RodrĂguez-Serrano, J.A., Perronnin, F.: A model-based sequence similarity with application to handwritten word spotting. IEEE Trans. Pattern Anal. Mach. Intell. 34(11), 2108–2120 (2012)
Perronnin, F., Rodriguez-Serrano, J.A.: Fisher kernels for handwritten word-spotting. In: International Conference on Document Analysis and Recognition, pp. 106–110 (2009)
Conte, D., Foggia, P., Sansone, C., Vento, M.: Thirty years of graph matching in pattern recognition. Int. J. Pattern Recognit. Artif. Intell. 18(03), 265–298 (2004)
Riesen, K.: Structural Pattern Recognition with Graph Edit Distance. Advances in Computer Vision and Pattern Recognition. Springer, Cham (2015). doi:10.1007/978-3-319-27252-8
Stauffer, M., Tschachtli, T., Fischer, A., Riesen, K.: A survey on applications of bipartite graph edit distance. In: Foggia, P., Liu, C.-L., Vento, M. (eds.) GbRPR 2017. LNCS, vol. 10310, pp. 242–252. Springer, Cham (2017). doi:10.1007/978-3-319-58961-9_22
Wang, P., Eglin, V., Garcia, C., Largeron, C., Llados, J., Fornes, A.: A novel learning-free word spotting approach based on graph representation. In: International Workshop on Document Analysis Systems, pp. 207–211 (2014)
Bui, Q.A., Visani, M., Mullot, R.: Unsupervised word spotting using a graph representation based on invariants. In: International Conference on Document Analysis and Recognition, pp. 616–620 (2015)
Riba, P., Llados, J., Fornes, A.: Handwritten word spotting by inexact matching of grapheme graphs. In: International Conference on Document Analysis and Recognition, pp. 781–785 (2015)
Stauffer, M., Fischer, A., Riesen, K.: Graph-based keyword spotting in historical handwritten documents. In: International Workshop on Structural, Syntactic, and Statistical Pattern Recognition (2016)
Stauffer, M., Fischer, A., Riesen, K.: A novel graph database for handwritten word images. In: International Workshop on Structural, Syntactic, and Statistical Pattern Recognition (2016)
Stauffer, M., Fischer, A., Riesen, K.: Speeding-up graph-based keyword spotting in historical handwritten documents. In: Foggia, P., Liu, C.-L., Vento, M. (eds.) GbRPR 2017. LNCS, vol. 10310, pp. 83–93. Springer, Cham (2017). doi:10.1007/978-3-319-58961-9_8
Bunke, H., Allermann, G.: Inexact graph matching for structural pattern recognition. Pattern Recognit. Lett. 1(4), 245–253 (1983)
Berretti, S., Del Bimbo, A., Vicario, E.: Efficient matching and indexing of graph models in content-based retrieval. IEEE Trans. Pattern Anal. Mach. Intell. 23(10), 1089–1105 (2001)
Fankhauser, S., Riesen, K., Bunke, H.: Speeding up graph edit distance computation through fast bipartite matching. In: Jiang, X., Ferrer, M., Torsello, A. (eds.) GbRPR 2011. LNCS, vol. 6658, pp. 102–111. Springer, Heidelberg (2011). doi:10.1007/978-3-642-20844-7_11
Koopmans, T.C., Beckmann, M.: Assignment problems and the location of economic activities. Econometrica 25(1), 53 (1957)
Riesen, K., Bunke, H.: Approximate graph edit distance computation by means of bipartite graph matching. Image Vis. Comput. 27(7), 950–959 (2009)
Burkard, R., Dell’Amico, M., Martello, S.: Assignment Problems (2009)
Acknowledgments
This work has been supported by the Hasler Foundation Switzerland.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Stauffer, M., Fischer, A., Riesen, K. (2017). Speeding-Up Graph-Based Keyword Spotting by Quadtree Segmentations. In: Felsberg, M., Heyden, A., KrĂĽger, N. (eds) Computer Analysis of Images and Patterns. CAIP 2017. Lecture Notes in Computer Science(), vol 10424. Springer, Cham. https://doi.org/10.1007/978-3-319-64689-3_25
Download citation
DOI: https://doi.org/10.1007/978-3-319-64689-3_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-64688-6
Online ISBN: 978-3-319-64689-3
eBook Packages: Computer ScienceComputer Science (R0)