Improved Two-Step Binarization of Degraded Document Images Based on Gaussian Mixture Model.
Robert KrupinskiPiotr LechKrzysztof OkarmaPublished in: ICCS (5) (2020)
Keyphrases
- document images
- gaussian mixture model
- ocr systems
- mixture model
- document image analysis
- expectation maximization
- document analysis
- em algorithm
- optical character recognition
- background subtraction
- maximum likelihood
- image binarization
- feature vectors
- feature space
- page segmentation
- speaker identification
- document processing
- printed documents
- language identification
- post processing
- word spotting
- gaussian mixture modeling
- text lines
- scanned documents
- prior knowledge
- machine learning