a9t9 / free-ocr-windows-desktop Goto Github PK

Im getting a warning when pressing the "Start OCR button" or the menu File>Start Ocr.

"Ooops. Test image not found (no big deal, everything else should still work fine).
I looked in C:\ProgramData(a9t9)OcrDesktop\tessdata"

The desktop app worked fine but I'm not getting any OCR results running the sourcecode.

Additional question: I'd like to add an option to export to .csv or .xlsx but as I'm a beginner I am still unsure how to add this option.
I assume it should be done in the FileManipulator.cs and |FileManipulator.cs

After installing your software, I miraculously have a folder C:\ProgramFiles\gs and my "Programs and Features" list has ghostscript in it.

If you need ghostscript, please include the binaries in your software and don't install gs as a separate software without the users consent. And why does my computer need a restart after installing your software? Did you also manipulate system files?

You are using elevated rights to perform actions, that I did not know about. I cannot recommend this software to anyone under those circumstances. 😠

I want to add Other language, download language form here:

https://github.com/tesseract-ocr/langdata

It is not possible to build the software without additional components.

Here are some of the warnings reported by Visual Studio 2017 (German):

Warnung		Das benutzerdefinierte Tool "XamlIntelliSenseFileGenerator" kann nicht gefunden werden.	(a9t9)OcrDesktop\App.xaml		
Warnung		Die Komponente "Ghostscript.NET", auf die verwiesen wird, wurde nicht gefunden.	(a9t9)OcrDesktop
Warnung		Die Komponente "GongSolutions.Wpf.DragDrop", auf die verwiesen wird, wurde nicht gefunden.	(a9t9)OcrDesktop
Warnung		Die Komponente "System.Windows.Interactivity", auf die verwiesen wird, wurde nicht gefunden.	(a9t9)OcrDesktop
Warnung		Die Komponente "Tesseract", auf die verwiesen wird, wurde nicht gefunden.	(a9t9)OcrDesktop

Where can I get the missing components and the missing tool?

Hi,

I couldn't have it working in my computer yet but just looked through your code.
It seems that it just uses Tesseract-OCR's basic function to detect texts. Which is:

using (var engine = new TesseractEngine(_pathToTestData, Language, EngineMode.Default))
{
using (var img = Pix.LoadFromFile(pathToImage))
{
using (var page = engine.Process(img))
{
var resultText = page.GetText();
if (!String.IsNullOrEmpty(resultText))
return resultText;
}
}
}

I know about this code and it wasn't very great. However your online OCR (https://ocr.space/) works really great. How come they are different? Are they really using same code?

I guess they are using different tessdata. Would you be able to share your tessdata too?