A Comparison of Some Morphological Filters for Improving OCR Performance

• Main Authors: Mennillo, Laurent, Cousty, Jean, Najman, Laurent
• Other Authors: Institut Pascal (IP), SIGMA Clermont (SIGMA Clermont)-Université Clermont Auvergne 2017-2020 (UCA 2017-2020 )-Centre National de la Recherche Scientifique (CNRS), Laboratoire d'Informatique Gaspard-Monge (LIGM), École des Ponts ParisTech (ENPC)-Centre National de la Recherche Scientifique (CNRS)-Université Gustave Eiffel, Benediktsson, J.A., Chanussot, J., Najman, L., Talbot, H.
• Format: Conference Object
• Language: English
• Published: HAL CCSD 2015
• Subjects: Character recognition; Morphological filtering; Vertex; Graphs; Simplicial complexes; [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV]; [INFO.INFO-TT]Computer Science [cs]/Document and Text Processing; Iceland
• Online Access: https://hal.science/hal-01168641; https://hal.science/hal-01168641/document; https://hal.science/hal-01168641/file/Report.pdf; https://doi.org/10.1007/978-3-319-18720-4_12

International audience Studying discrete space representations has recently lead to the development of novel morphological operators. To date, there has been no study evaluating the performances of those novel operators with respect to a specific application. This article compares the capability of several morphological operators, both old and new, to improve OCR performance when used as preprocessing filters. We design an experiment using the Tesseract OCR engine on binary images degraded with a realistic document-dedicated noise model. We assess the performances of some morphological filters acting in complex, graph and vertex spaces, including the area filters. This experiment reveals the good overall performance of complex and graph filters. MSE measures have also been performed to evaluate the denoising capability of these filters, which again confirms the performances of both complex and graph filtering on this aspect.