Segmentation and classification for mixed text/image documents using neural network.
Shinichi ImadeSeiji TatsutaToshiaki WadaPublished in: ICDAR (1993)
Keyphrases
- segmentation method
- neural network
- image classification
- image segmentation
- segmentation algorithm
- image analysis
- multiscale
- grey level
- pixel classification
- document classification
- text lines
- test images
- image regions
- document categorization
- image data
- text documents
- textural features
- energy function
- input image
- web documents
- scanned documents
- image representation
- segmented images
- free text
- text information
- information retrieval
- image content
- edge detection
- automatic categorization
- topic segmentation
- printed documents
- text classifiers
- text mining
- bounding box
- pixel level
- web images
- line extraction
- document images
- textual information
- image features
- text classification
- text retrieval
- document analysis
- complex background
- document collections
- document clustering
- feature vectors
- image retrieval
- handwritten text
- keywords
- image segments
- energy functional
- grey level co occurrence matrix
- text queries
- shape prior
- digital documents
- handwritten documents
- segmented regions
- multimedia documents
- foreground and background