Aspose OCR don´t recognize Text into Image

Hello;

I have this code to make OCR recognition with the last version of ASPOSE OCR (2.7.0). The image is generated with 300ppp

Aspose.OCR.License license = new License();
license.SetLicense(“Aspose.Total.lic”);
// The path to the documents directory.

//Initialize an instance of OcrEngine
OcrEngine ocrEngine = new OcrEngine();

//Set the Image property by loading the image from file path location or an instance of MemoryStream
ocrEngine.Image = ImageStream.FromFile(“file.tif”);

//Clear the default language (English)
ocrEngine.LanguageContainer.Clear();

//Load the resources of the language from file path location or an instance of Stream
ocrEngine.LanguageContainer.AddLanguage(LanguageFactory.Load(“Aspose.OCR.Spanish.Resources.zip”));

//Process the image
if (ocrEngine.Process())
{
//Display the recognized text
Console.WriteLine(ocrEngine.Text);
System.IO.StreamWriter file = new System.IO.StreamWriter(“d:\test2.txt”);
file.Write(ocrEngine.Text);

file.Close();
}

This code generate unreadable text. More or less like this:

ToppWs@$%%%%%@p*pp’00%$3VZ$$$%$$#%=$cv’nmVMWJ-V50V%@Ii-lyii-ininNoN$!c#N=39%NpN###.NNNININN.ppNNNONc#\NpppN$.NZFNNNNWNNNNNNNNNNNNWNNNNNNJNONNNNNNNNNWNNNN.ppphp%NNN#JNNNNNNNNNNNUMNNNNNUMMNNNWNNNNNWNW90%%%%J##@#p’’’’’’’’’’’‘3’’’’’’’’’’’’’’’’’’‘m’’’’’‘V’’’’‘0’‘V’’’’‘y’’’’’’’’’’’’’’‘9’’’’‘a’‘NNNNNPNNNNpNp#N$$$#N#p##N#W$###p#$%$#9#J#N###NNUNNNNWNNNWNJ9NpNNNNNNNWNNW#=9NNWN#999NNNNNNNNNNN.NNWNNNNNNNNNNNN!NNNNNN%N#p#NaNNNWNNOVV—‘w’’’’’‘0’’’’’‘a’’‘y’%’’’’‘w’’’‘a’‘i’’‘g’’’’’’’‘g’’’’’’‘m’’’’’’‘a’’’'win%v%$JT’empZ#J#$%'hp%%%%%%NNJs$'p’00N$9999’phpp’NF37NevyW$N#N#=59%3$J$35%#NNNWMNpphphp#%NNFTWONNNUJJ9999999999999%%%%%%‘ppW%$%$50%N#’$99%#$#$%#%pN$$$$30JNWFW950N$'pp$%$'phpW$39%J####33NJUNNWM#$NN3NW9NNN3#W’9990999999%9900N%'pphp#NWNWNNMMUNMVNZ$9Np$p’3Newp9N%Z$NZ$$#‘hp%############$N$9N$$N######$%#$%$5NNNMMMNNW’$$90%N$9999999905%79’00%99970pphp33%9#########%#J#%NJOJ#9#SWFN#990093%%=JuN$##p####'pphp%N#$'pph%%p####‘WNWWWW’ppN$Z##$N$%$$n’‘hm$%$$NNUMNUJWomphp%%%%N$33%NN’$909%%%N$$%#cipp’NWNWNW9999%%%%%%%%%N%999979N%J#V\V%$$V`

I am working with a .net 4.5 platform.

Thank you

Hi,


Thank you for the sample image. We are currently evaluating the presented scenario against the latest version of Aspose.OCR for .NET 2.7.0, and we will shortly get back to you with updates in this regard.

Hi again,


This is to inform you that we were not able to get the desired results while using the latest version of Aspose.OCR for .NET 2.7.0 against your provided sample therefore we have logged this incident in our bug tracking system under the ticket OCR-34095 for product team’s review. Please allow us some time to properly analyze the problem cause, and get back to you with updates in this regard.

We are sorry for the inconvenience.

Ok.
Thank you.


Some news?

Hi,

Thank you for your inquiry.

This is to update you that our product team has analyzed the issue and identified the cause of the problem. They are now working on changing the algorithm for textblocks detection so that scenarios like the one mentioned by you can be handled appropriately. This process definitely takes some time. Please allow us some more time. We will update you accordingly.

We are sorry for the inconvenience caused.

Hello

Do you have some news about this issue?
I have seen that a new version of ASpose OCR has released, this new version resolves our problem?

Thank you

Hi,

Thank you for contacting us back.

I am afraid, the ticket OCR-34095 is currently unresolved therefore the fix was not included in the recently published release. We have requested the product team to provide the insight of the the problem and preferably an estimated release schedule for the fix. As soon as we receive any updates, we will post here for your kind reference.

Hi again,

We have discussed the problem (logged earlier as OCR-34095) with the product team. The reason behind the low accuracy rate of the recognized text is the incorrect layout detection and low character recognition quality. In order to provide the fix for the aforesaid ticket, we have to make fixes to the text block detection algorithm and also improve the character detection. We have already started work on the aforesaid issues by attaching the blocking tasks, however, due to the required amount of work we may not be able to provide the fix very soon. As soon as we have made significant progress in achieving the desired goals, we will notify you here.

Thank you for your patience with us.

Hello,

Some news about this issue?

Regards

Hi,

This issue is blocked by several sub-issues. Some of these issues have been fixed but most of them are still unresolved. This issue will be fixed once all sub-issues are fixed.

The issue is expected to be fixed somewhere in 2016 however we are not in a position to share a concrete ETA at the moment. We will update you as soon as the issue is fixed or we have a concrete ETA for this issue.

Sorry for the inconvenience.

Best Regards,

The issues you have found earlier (filed as ) have been fixed in this Aspose.Words for JasperReports 18.3 update.