Document Image Processing

© 2019 by the authors; CC BY license

synthetic image generation; document degradation models; performance evaluation; data augmentation for retraining and fine-tuning; DIAR; Arabic OCR systems; holistic OCR approach; holistic OCR features; lexicon reduction; historical handwritten transcription; out-of-vocabulary word recognition; character-level language model; word structure retrieval; documents; binarization; back-to-front interference; bleeding; video text detection; video text recognition; AcTiV dataset; Arabic Video OCR; DTW distance; query classifiers; word spotting; indexing; retrieval; Classifier combination; Dempster-Shafer theory of evidence; Indic script identification; Histograms of Oriented Gradients; Modified Log-Gabor filter transform; Elliptical features; handwritten character recognition; deep learning; Devanagari characters; convolutional neural network; adaptive gradient methods; document image analysis; binarization; character recognition; text line segmentation; word recognition; transliteration; palm leaf manuscript; dataset; benchmark; experimental test; text/non-text separation; local binary pattern; handwritten document; document image processing; texture-based features; ancient document restoration; image inpainting; bleed-through removal; sparse representation; slant removal; document image processing; document image page; document image processing; preprocessing; document restoration; binarization; slant removal; text-line segmentation; handwriting recognition; indic/arabic/asian scripts; OCR; Video OCR; word spotting; retrieval; document datasets; performance evaluation; document annotation tools