Free open-source OCR application for the Windows Store - A modern GUI front-end for the Microsoft OCR library. The application also includes support for reading and OCR'ing PDF files.
I OCR'd an image file - not a bad recognition and saved to text file - file was empty
Did the same again and saved to Word document - another empty file.
I cannot upload the file produced as it is empty.
Files attached show the process I went through.
I tried opening the solution using VS2017 Community and it gave me an error message that the project is not compatible with this version of VS. I have my VS and win10 updated to latest. it also opened this web page.
When submitting a smartphone image of bilingual german/japanese text to the web OCR service set to "Japanese", the german parts are recognized almost flawlessly, but none of the chinese or japanese characters are. This does not seem to be an issue of the OCR engine since after selecting "Chinese Traditional", the chinese characters are recognized very well (but (obviously) the german is garbled and japanese hiragana are still not recognized).
Hi,
for bilingual texts (say, language textbooks) a dual-language mode would be great. If necessary, I can provide some german/japanese and english/japanese test data.
I posted this to your blog, but I think Github is easier to work with, so just ignore my previous inquiries.
I noticed that Microsoft's Project Oxford OCR offers an "Auto Detect" language option. I assume it uses the Microsoft OCR Engine. Can you enable this in your software?
I would like to use the library in an offline embedded application. Is it possible to compile the Microsoft OCR Engine into an executable with a command line interface? So that we could run for example, msocr.exe -i input.jpg -o output.json.