Incorporating Conditional Independence Assumption with Support Vector Machines to Enhance Handwritten Character Segmentation Performance.
Manolis MaragoudakisErgina KavallieratouNikos FakotakisPublished in: ICPR (3) (2002)
Keyphrases
- character segmentation
- character recognition
- automatic recognition
- optical character recognition
- conditional independence assumption
- gray scale images
- hand written
- naive bayes
- license plate
- attribute dependencies
- data sets
- machine vision
- segmentation method
- printed documents
- gray scale
- color images
- training data
- document images
- image processing
- test instances
- handwriting recognition
- learning algorithm
- binary images