First Sentence:
The process of generating Character Shape Codes (CSCs) encodes only the grossest features of character image size and placement, while ignoring the high spatial frequency features that Optical Character Recognition (OCR) relies upon [1][2].
Read the first page
Key Phrases - Statistically Improbable Phrases (SIPs):
(learn more)
vertical text lines, amount recognizer, lexical search approach, hierarchical layout structure, legal amount recognition, shape coding process, subregion candidates, cumulative recognition rate, font style detection, possible cut area, text line location, lexicon tree, correct segmentation rate, courtesy amount recognition, wavelet reconstruction algorithm, order hold interpolation, form document image, average character height, garbage detection, endpoint features, separator elements, document analysis systems, projection blocks, coincident area, character classifier
Key Phrases - Capitalized Phrases (CAPs):
(learn more)
Pattern Recognition, International Conference, Add-On Code, Las Vegas, Computer Science, San Jose, Hidden Markov Models, New York, World Scientific, Analytical Word Recognizer, Graphite Light, Information Science Research Institute, Lecture Notes, Times New Roman, Document Management System, Script Recognizer, United States, University of Nevada, University of Washington, Holistic Word Recognizer, Japan Abstract, Object-Process Diagrams, Automatic Document Processing, Central Research Laboratory, Lockheed Martin Postal Systems
New!
Books on Related Topics |
Concordance
|
Text Stats
Browse Sample Pages:
Front Cover |
Table of Contents |
First Pages |
Back Cover |
Surprise Me!