Closed JafferSheriff closed 9 years ago
@JafferSheriff, pls, refer to Tips for Improving OCR Results Let me know if that could help you.
Guys I tried with GPUImage's Adaptive Threshold Filter and bypassed Tesseract's Internal Thresholder but still not getting acceptable output with tessaract . I took a pic of text "Stack Overflow" from their website and applied GPUImage's Adaptive Threshold Filter with tessaract Internal Thresholder disabled and I got recognized text as "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n" .I have configured everything just like the way they instructed and also added tessdata which contain bunch of files for english language recognizion . Seriously I'm not able to make it to work to get atleast acceptable level of output. So help me guys .
@ws233 What might be the problem ?
From here: Tesseract performs OCR best when it is given a preprocessed image that is ideally crystal clear black text on a pure white background.
Your image from the screenshot above doesn't seem ideally crystal clear black text on a pure white background. Use some image processing techniques.
Thank you @ws233 for your valuable suggestion.
Why Tesseract-OCR-iOS is not all recognizing ocr. I tried with image taken from camera in iphone 5s 9.0.2 but not even a single character is correctly recognized. CamScanner Ios app does this job pretty much amazingly and why don't you just try to process the image to remove unwanted information using core graphics and apply that to recognizer. Is there any way ocr possible in ios with camera ?