How Can I Train Tesseract to recognize handwriting?

I have tesseract installed via cocoa pods. I made an Xcode project using swift code that uses tesseractOCR to take a picture, then take the text out of the picture and make it stand alone. (I followed this tutorial: https://www.raywenderlich.com/2010498-tesseract-ocr-tutorial-for-ios) I was researching how I can train tesseract to recognize handwriting so my app can recognize handwritten text also, but I wasn’t getting a clear answer… If anyone that has done this in the past can tell me how I can modify or train tesseract to recognize handwriting without disrupting my previous commands please share!

Hi @slushypuffin, from my understanding of the tutorial that you followed, Tesseract has limitations in regards to recognizing handwriting unfortunately. It does not seem like Tesseract is ideal for recognizing handwriting however, I found this link that could be a door way to training Tesseract so it can recognize handwriting.

Best,
Gina

@gdelarosa I have visited that site before and I’m fairly new to coding so not all of it made sense… so correct me if I’m wrong: I need to download an handwriting image library (like IAImages) and then turn the files into box/tiff files so Tesseract can understand it? I’m not completely clear on how I go about converting the files and uploading them to trained data… I would appreciate some help as I’m sure you have experience in this field, Thanks!

Hi @slushypuffin, I’ve never used Tesseract but it seems like you are on the right track after briefly reading through the “Overview of Training Process”.

This topic was automatically closed after 166 days. New replies are no longer allowed.