Reranking with Linguistic and Semantic Features for Arabic Optical Character Recognition.
Nadi TomehNizar HabashRyan RothNoura FarraPradeep DasigiMona T. DiabPublished in: ACL (2) (2013)
Keyphrases
- optical character recognition
- semantic features
- linguistic features
- visual features
- ocr systems
- character recognition
- document images
- semantic information
- low level features
- wordnet
- text classification
- image search
- structural features
- handwriting recognition
- printed documents
- image classification
- semantic similarity
- visual information
- feature set
- document clustering
- image retrieval
- scanned documents
- natural language processing
- low level
- word spotting
- printed text
- natural language
- image annotation
- higher level
- domain knowledge
- handwritten documents
- feature extraction
- high level
- image processing
- feature selection