Word Search in Handwritten Text Based on Stroke Segmentation

Handwritten archival documents form a fundamental part of humanity's cultural heritage. However, their analysis remains a labor-intensive task for professional researchers, such as historians, philologists, and linguists. Unlike commercial OCR applications, working with historical manuscripts requires a fundamentally different approach due to the extreme diversity of handwriting, the presence of corrections, and material degradation.


This paper proposes a method for searching within handwritten texts based on stroke segmentation. Instead of performing full text recognition, which is often unattainable for historical documents, this method allows for efficiently answering researcher search queries. The key idea involves decomposing the text into elementary strokes, forming semantic vector representations using contrastive learning, followed by clustering and classification to create an adaptive handwriting dictionary.


It is experimentally shown that search by comparing tuples of reduced sequences of the most informative strokes using the Levenshtein distance provides sufficient quality for the task at hand. The method demonstrates resilience to individual handwriting characteristics and writing variations, which is particularly important for working with authors' archives and historical documents.


The proposed approach opens up new possibilities for accelerating scientific research in the humanities, reducing the time required to find relevant information from weeks to minutes, thereby qualitatively transforming research capabilities when working with large archives of handwritten documents.

handwritten text, search, stroke analysis, segmentation, vector representation, contrastive learning, clustering

1. Zhang X.-Y., Sun Z., Jin L., Ni H. & Lyons T. J. Learning Spatial–Semantic Context with Fully Convolutional Recurrent Network for Online Handwritten Chinese Text Recognition // IEEE Transactions on Pattern Analysis and Machine Intelligence. 2018. Vol. 40, no. 8. P. 1903–1917. https://doi.org/10.1109/tpami.2017.2732978

2. Rahal N., Vögtlin L., Ingold R. Historical Document Image Analysis Using Controlled Data for Pretraining // International Journal on Document Analysis and Recognition (IJDAR). 2023. Vol. 26, no. 3. P. 241–254. https://doi.org/10.1007/s10032-023-00437-8

3. Puigcerver J. Are Multidimensional Recurrent Layers Really Necessary for Handwritten Text Recognition? // 14th IAPR International Conference on Document Analysis and Recognition (ICDAR). 2017. Vol. 1. P. 67–72. https://doi.org/10.1109/ICDAR.2017.20

4. Rath T. M., Manmatha R. Word Spotting for Historical Documents // International Journal on Document Analysis and Recognition (IJDAR). 2007. Vol. 9, no. 2–4. P. 139–152. https://doi.org/10.1007/s10032-006-0027-8

5. Mestetskii L.M. Continuous Morphology of Binary Images: Figures, Skeletons, Circulars. M.: FIZMATLIT, 2009. 231 p.

6. Mestetskiy L.M. Stroke Segmentation of Handwritten Text Based on Medial Representation // Pattern Recognition and Image Analysis: Advances in Mathematical Theory and Applications. 2024. Vol. 34, no. 4. P. 1185-1191. https://doi.org/10.1134/S1054661824701256

7. Dias C. da S., Britto Jr. A. de S., Barddal J. P., Heutte L., Koerich A. L. Pattern Spotting and Image Retrieval in Historical Documents using Deep Hashing. 2022. arXiv:2208.02397

8. He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition // IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016. P. 770–778. https://doi.org/10.1109/CVPR.2016.90

9. Ester M., Kriegel H.-P., Sander J., Xu X. A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise // 2nd International Conference on Knowledge Discovery and Data Mining (KDD). 1996. P. 226–231.