Lead Tools 16.5 OCR speed

I'm just enquiring as to whether or not there are any options that i can switch on or off to speed up the OCR of a tif file. I've noticed in the
Private Sub OCRRecognizeCallback(ByVal data As IOcrProgressData)
that the data.Operation says it goes through 3 passes on each page... however it never finds anything different on the 2nd and 3rd pass can i stop it from going to the 2nd and 3rd pass... also what about the spell checking portion can i turn that off?

thanks

i found how to turn on the spell check
_ocrEngine.SpellCheckManager.Enabled = False

The recognition speed depends on the image that you are trying to OCR. However, you may try to reduce the recognition time by changing the Recognition module Tradeoff to Accurate, Balanced or Fast.
- Accurate: Most accurate recognition.
- Balanced: Mid-level accuracy/speed recognition
- Fast: Fast recognition
For more information, please see the help topic "LEADTOOLS OCR Professional Engine Settings" in the LEADTOOLS .Net documentation.

Also, you mentioned that "data.Operation says it goes through 3 passes on each page". What exactly do you mean by passes? Please provide me with more details about this issue?

Thanks,
Maen Badwan
LEADTOOLS Technical Support

i found the documentation for tradeoff thank you...
ocrDocument.Engine.SettingManager.SetEnumValue("Recognition.RecognitionModuleTradeoff", "Fast")

Also as for the passes i just assumed it was because i have set up a procedure
Private Sub OCRRecognizeCallback(ByVal data As IOcrProgressData)

and in that proc i print out
data.Operation.ToString()

it does hit this sub many many times during the ocr but the messages stay consistant

data.Operation.ToString() outputs
"RecognizeSecondPass"
AND
"RecognizeThirdPass"

Normal
0


false
false
false







MicrosoftInternetExplorer4








/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:10.0pt;
font-family:"Times New Roman";
mso-ansi-language:#0400;
mso-fareast-language:#0400;
mso-bidi-language:#0400;}

The
number of passes depends on Recognition Module you use. For example, if you
enable OcrZoneRecognitionModule.OmniFontPlus3WayVoting, you will get 3 passes,
and the OCR engine will perform voting between the 3 passes to get an accurate result.

do you have an example of how to set that?

We do not have a ready sample for this. However, you can know which recognition modules are supported by the OCR engine on your machine using the GetSupportedRecognitionModules Method. If the "OmniFontPlus3WayVoting" module is supported, you can use it in your code using the OcrZone.RecognitionModule property.

i don't need a full on example just one or 2 lines would suffice i can't seem to quite figure out how to set it up

If you want to turn off OmniFontPlus3WayVoting Recognition Module,
Simply set OcrZone.RecognitionModule to OcrZoneRecognitionModule.OmniFontMor.

When you use the MOR module alone without the 2-way or the 3-way voting module, you only get one pass. Please try the following code:
+----------+
OcrZoneRecognitionModule[] recognitionModules = _ocrEngine.ZoneManager.GetSupportedRecognitionModules();
foreach (OcrZoneRecognitionModule recognitionModule in recognitionModules)
if (recognitionModule == OcrZoneRecognitionModule.OmniFontMor)
{
using (IOcrDocument ocrDocument = _ocrEngine.DocumentManager.CreateDocument())
{
// Add a page to the document                        
IOcrPage ocrPage = ocrDocument.Pages.AddPage(tifFileName, null);
                       
ocrPage.AutoZone(null);

OcrZone Zone = ocrPage.Zones[0];
Zone.RecognitionModule = OcrZoneRecognitionModule.OmniFontMor;
}                  
}
+----------+

Thanks,
Maen Badwan
LEADTOOLS Technical Support