HairyFotr/OCRTrain
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Repository files navigation
------------- OCR Trainer ------------- 1. Put your images into img/ 2. Put your transcriptions into a text file... "filename transcription" 3. Run runTrain <yourTextFile> 4. Wait and wait and wait :) Things you need: sudo apt-get install imagemagick tesseract-ocr gocr cuneiform ocrad Things you should probably set (a.k.a. things I should abstract away into files): allowedCharacters / stringFilter the appropriate string scoring algorithm different sequence of param-changing algorithms