Freebie OCR for screenshots and TIFF/BMP Files

[gpulawski]

I looked into using a couple of the commercial SDK's for some limited OCR of screenshot bitmaps for instance to verify information in a window and to check some image files for certain text content. These were quite expensive and required royalty fees. Then, I recalled that Microsoft Office has built in OCR. It is in Office Tools for version 2003 onward -- in Microsoft Office Document Imaging. START>ALL PROGRAMS>MICROSOFT OFFICE>MICROSOFT OFFICE TOOLS>. The component is called MODI and lives in MDIVWCTL.DLL. It is scriptable!!!

MODI understands TIFF, multi-page TIFF, and BMP. You can search MSDN under MODI for objects and methods.

Here's some example code to open a TIF or BMP file, OCR the first page, and then put the text result into a single string in a message box.

VBStart Dim miDoc Dim miLayout Dim stringOut set miDoc=CreateObject("MODI.Document") ' Load an existing TIFF file. miDoc.Create ("C:\pathname.tif") ' Perform OCR. 'You can change the mousepointer here to an hourglass or something. miDoc.Images(0).OCR 'Change the mouse back to normal default. set miLayout = miDoc.Images(0).Layout stringOut=miLayout.Text MsgBox(stringOut) Set miLayout = Nothing Set miDoc = Nothing VBEND

MODI also has its own search methods, document viewer, etc. I think Office 2007 and Vista do not install it by default like 2003 does. (I don't own Vista.)

[mtettmar]

Hi,

Thanks for this, this is awesome. MODI (Microsoft Office Document Imaging) is installed with my Office 2007 installation under Vista.

Here's my version of your code that uses OCR to get the text of the active window. Uncomment the comment block to OCR the entire screen:

VBStart Function DoOCR(bitmapfile) Dim miDoc Dim miLayout Dim stringOut set miDoc=CreateObject("MODI.Document") miDoc.Create (bitmapfile) ' Perform OCR. 'You can change the mousepointer here to an hourglass or something. miDoc.Images(0).OCR 'Change the mouse back to normal default. set miLayout = miDoc.Images(0).Layout stringOut=miLayout.Text 'MsgBox(stringOut) DoOCR = stringOut Set miLayout = Nothing Set miDoc = Nothing End Function VBEND /* //This example OCRs entire screen //Capture Entire Screen to bitmap GetScreenRes>x,y ScreenCapture>0,0,x,y,%TEMP_DIR%\~scrn.bmp //Use OCR to get text from screen VBEval>DoOCR("%TEMP_DIR%\~scrn.bmp"),TheText //Display text MessageModal>TheText */ //This example OCRs just the active window GetActiveWindow>title,X,Y,W,H ScreenCapture>x,Y,{%X%+%W%},{%Y%+%H%},%TEMP_DIR%\~scrn.bmp //Use OCR to get text from active window VBEval>DoOCR("%TEMP_DIR%\~scrn.bmp"),TheText //Display the text MessageModal>TheText

_________________
Regards,
Marcus Tettmar
http://mjtnet.com/blog/ | http://twitter.com/marcustettmar
Please do not email/PM me for private support - post to the forum so that everyone benefits. For private support please send email via our web site.

[gpulawski]

Yup, I can think of a lot of possibilities. It doesn't compare in accuracy to Omnipage or ABBYY but it's good enough for a lot of interesting Automation/MacroScheduler ideas. (And just about everybody has it installed already.)

GJP

[pzelenka]

This is very cool.

I used this in a script to process Windows exception dialog boxes with text that was otherwise impossible to get. When the exception text is evaluated, the script can then be appropriately branched.

Thank you very much!!

[mtettmar]

Just in case you weren't aware - with most windows message boxes if you hit CTRL+C the content of the dialog is copied to the clipboard.
_________________
Regards,
Marcus Tettmar
http://mjtnet.com/blog/ | http://twitter.com/marcustettmar
Please do not email/PM me for private support - post to the forum so that everyone benefits. For private support please send email via our web site.

[pzelenka]

I gave the CTRL+C a try, also efforts using GetWindowText and GetControlText approaches. all without success.

When there are exception issues, you will be presented with warning dialog boxes with differing descriptions, usually dependent upon the module, routine or dictionary. The most pressing concern is the proper handling of the Warning dialog boxes.

In the case of the Dialog boxes, the window title is available, but the descriptive content was out of reach, until I used the screen capture and OCR approach.

Thanks

[Semper]

Hi,

any idea why i can't run the MODI script wihtout
getting error OCR:bad language at
line miDoc.Images(0).OCR

Thanks for the feedback.

[Semper]

I figured it out. The language parameter Id was missing since i'm on a
multilingual machine

miDoc.images(0).OCR(9) did the job.

[rullbandspelare]

I dont have Office 2007 and was looking for OCR. I found this one wich is free to use.
http://www.topocr.com/
It would be interesting if someone could compare the quality of result with the previously posted example for Office 2007.

for example:

GetActiveWindow>title,X,Y,W,H ScreenCapture>x,Y,{%X%+%W%},{%Y%+%H%},screen.jpg run program>topocr.exe screen.jpg screen.txt ReadFile>screen.txt,screen MessageModal>%screen%

I was also looking for Barcode reader OCR. And found this:
http://www.metois.com/Eymbarcode/index.htm
This free demo gives a splash screen every now and then but works very well.
Has someone found anything better?

Thanks!