Color displays and color science
Color and the computer
Locating and Recognizing Text in WWW Images
Information Retrieval
Extracting Text from WWW Images
ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
Fuzzy Segmentation of Characters in Web Images Based on Human Colour Perception
DAS '02 Proceedings of the 5th International Workshop on Document Analysis Systems V
3D mesh simplification techniques for image-page clusters detection
WSEAS Transactions on Information Science and Applications
Hi-index | 0.00 |
There is a significant need to recognise the text inimages on web pages, both for effective indexing and forpresentation by non-visual means (e.g., audio). Thispaper presents and compares two novel methods for thesegmentation of characters for subsequent extraction andrecognition. The novelty of both approaches is thecombination of (different in each case) topologicalfeatures of characters with an anthropocentricperspective of colour perception- in preference to RGBspace analysis. Both approaches enable the extraction oftext in complex situations such as in the presence ofvarying colour and texture (characters and background).