Get image data

4

My system must process some images and get key data in text. Of course, the images will follow a style pattern.

Where can I start? Is there any API that can help me?

    
asked by anonymous 26.06.2015 / 21:40

1 answer

4

The most common and free OCR (character recognition) engine I know is Tesseract. There are two contributions of it to .NET:

I recommend passing the images to black and white before using Tesseract to read. Preferably, try to position the texts horizontally, and do not pass very small texts, which can be lost.

Another tip is to set up a list of allowed characters to read . Depending on how the text is, it can change some characters for others.

    
26.06.2015 / 21:53