Open Shreeshrii opened 6 years ago
@shreeshrii & @topherseance: there are more than 20 Javanese script fonts available here: https://bennylin.github.io/keyboards/jawa-fonts.html
@bennylin Are these Unicode fonts?
Yes
Are there any labelled datasets with scanned images and their Unicode groundtruth transcription that can be used for training/testing tesseract's jav-java traineddata?
What accuracy did the UKDW ocr achieve?
I'm not in the loop for the research. You might want to contact Dr. Lucia Krisnawati for that.
Originally posted in forum
https://groups.google.com/forum/?utm_medium=email&utm_source=footer#!msg/tesseract-ocr/8r8YOQgTBT4/xHpCTp9DAwAJ