Document Image Segmentation into Text, Tontinuous-tone and Screened-halftone Region by the Neural Networks.
Kazuaki NakamuraShinji YamamotoTetsuya ItohPublished in: PICS (1998)
Keyphrases
- neural network
- image segmentation
- text documents
- information retrieval
- digital documents
- web documents
- keywords
- document processing
- document analysis
- text content
- region segmentation
- boundary information
- textual content
- document content
- document images
- segmented images
- text collections
- printed documents
- scientific papers
- watershed transformation
- multimedia documents
- semantic information
- document collections
- document corpus
- markov random field
- automatic text summarization
- text corpus
- scanned documents
- scientific documents
- document representation
- document structure
- artificial neural networks
- information retrieval systems
- graph cuts
- segmented regions
- electronic documents
- automatic summarization
- halftoning algorithm
- computer vision
- text summarization
- multiscale
- text categorization
- document set
- gray scale
- gray level
- image processing
- text mining
- level set
- related documents
- handwritten documents
- image regions
- text retrieval
- document level
- text lines
- latent semantic analysis
- segmentation method
- segmentation algorithm
- active contours
- text classification
- region growing
- edge detection
- color images
- level set method
- web pages