The ICDAR 2021 Time-Quality Binarization Competition assessed the performance of 12 new and 49 other previously published binarization algorithms for scanned document images. Four test sets of “real-world” documents with different features were used. For each test set, the top twenty algorithms in the quality of the resulting two-tone images had their average processing time presented, yielding an account of their time complexity.
This is an author-produced, peer-reviewed version of this conference proceeding. The final, definitive version of this document can be found online at ICDAR 2021: Document Analysis and Recognition, published by Springer. Copyright restrictions may apply. https://doi.org/10.1007/978-3-030-86337-1_47
Dueire Lins, Rafael; Barros Bernardino, Rodrigo; Barney Smith, Elisa; and Kavallieratou, Ergina. (2021). "ICDAR 2021 Competition on Time-Quality Document Image Binarization". In J. Lladós, D. Lopresti, and S. Uchida (Eds.), ICDAR 2021: Document Analysis and Recognition (Lecture Notes in Computer Science series, Volume 12824, pp. 708-722). Springer. https://doi.org/10.1007/978-3-030-86337-1_47