Skip to content
/ my-ocr Public

Optical character recognition for english uppercase letters, lowercase letters, and digits 0 to 9.

License

Notifications You must be signed in to change notification settings

BH4/my-ocr

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 

Repository files navigation

my-ocr

Optical character recognition for English uppercase letters, lowercase letters, and digits 0 to 9.

Using "NIST Handprinted Forms and Characters Database" 2nd edition.

Current models are able to reach around 90% accuracy on a merged model in which some of the classes are combined into merged classes. The following list of tuples shows which classes are merged. Each of these merged classes includes an uppercase letter and its lowercase equivalent with the only addition being that the class containing the letter O also contains the number zero.

('c', 'C'), ('k', 'K'), ('m', 'M'), ('p', 'P'), ('s', 'S'), ('u', 'U'), ('v', 'V'), ('w', 'W'), ('x', 'X'), ('y', 'Y'), ('z', 'Z'), ('o', 'O', '0')

About

Optical character recognition for english uppercase letters, lowercase letters, and digits 0 to 9.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages