Document Image Restoration for OCR & Archives

This work presents an adaptive document restoration framework integrating stroke-aware en- hancement and OCR-guided evaluation. The system dynamically adapts to document quality variations and preserves textual structures during enhancement. Experimental results demonstrate improved readability and recognition performance, making the approach suitable for archival digitization and intelligent document processing systems.

[1]     N. Otsu, “A Threshold Selection Method from Gray-Level Histograms,” IEEE Transac- tions on Systems, Man, and Cybernetics, vol. 9, no. 1, pp. 62–66, 1979. [2]     J. Sauvola and M. Pietika¨inen, “Adaptive Document Image Binarization,” Pattern Recog- nition, vol. 33, no. 2, pp. 225–236, 2000. [3]     C. Wolf and J. Jolion, “Extraction and Recognition of Artificial Text in Multimedia Doc- uments,” Pattern Analysis and Applications, vol. 6, no. 4, pp. 309–326, 2004. [4]     I. Pratikakis, B. Gatos, and K. Ntirogiannis, “ICDAR Document Image Binarization Con- test (DIBCO),” International Conference on Document Analysis and Recognition (IC- DAR), pp. 1506–1510, 2009. [5]     N. R. Howe, “Document Binarization with Automatic Parameter Tuning,” International Journal on Document Analysis and Recognition, vol. 16, no. 3, pp. 247–258, 2013. [6]     O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomed- ical Image Segmentation,” International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 234–241, 2015. [7]     R. Smith, “An Overview of the Tesseract OCR Engine,” International Conference on Document Analysis and Recognition (ICDAR), pp. 629–633, 2007. [8]     D. Kang, S. Kim, and H. Lee, “Deep Learning-Based Document Image Enhancement for OCR Accuracy Improvement,” IEEE Access, vol. 8, pp. 162593–162603, 2020. [9]     L. Xu, C. Lu, Y. Xu, and J. Jia, “Image Smoothing via L0 Gradient Minimization,” ACM Transactions on Graphics, vol. 30, no. 6, pp. 1–12, 2011. [10]     B. Su, S. Lu, and C. L. Tan, “Robust Document Image Binarization Technique for De- graded Document Images,” IEEE Transactions on Image Processing, vol. 22, no. 4, pp. 1408–1417, 2013. [11]     M. Li, T. Lv, J. Cui, et al., “TrOCR: Transformer-Based Optical Character Recogni- tion with Pretrained Models,” Proceedings of AAAI Conference on Artificial Intelligence, 2023. [12]     S. Kumar and M. Singh, “A Survey of Document Image Enhancement Techniques for OCR Systems,” Journal of Visual Communication and Image Representation, vol. 58, pp. 536–548, 2019.