It is presented herein a new entropy-based segmentation algorithm for images of historical documents. The algorithm provides high quality images and it also improves OCR (Optical Character Recognition) responses for typed documents. It adapts its settings to achieve better quality images through changes in the logarithmic base that defines entropy. For this purpose, a measure for image fidelity is applied just as information inherent to images of documents. © Springer-Verlag 2004.
CITATION STYLE
De Mello, C. A. B. (2004). Image segmentation of historical documents: Using a quality index. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3212, 209–216. https://doi.org/10.1007/978-3-540-30126-4_26
Mendeley helps you to discover research relevant for your work.