Needed Aspose.OCR

Aspose.OCR Product Family
1

We need something to do zonal OCR, a rich and solid API like the other products that you guys have been making.

I am impressed by the usefulness of the Aspose products to date, I have an Aspose.Total subscription for a few years new and I have not regretted it not a single day !!

Lets get something to give ABBYY and the others a run for their money ...

2

Dear Daniel,

Thank you for your request.

We have thought about this product. Once we have some promising results we will keep you posted. On the other hand, if you know some one(s) good at this area, feel free to let me know.

3

Hello,

I too am interested in an Aspose OCR package.

Regards,
JP O’Gorman

4

hi,

i was wondering if aspose has the following capabilities:

1. ocr in java an image and produce the extracted text.
2. ocr in java an image and produce the extracted barcodes.

This message was posted using Aspose.Live 2 Forum
5

Hi,

You can encode binary files as well as text and create the barcode as an image file. And you can reverse that process i.e. extract the barcode from the image and reproduce the original binary file or text.

What are your specific requirements? We would be able to guide/recommend you in detail.

6

The specific requirements are as I mentioned in my original post, more or less:

1. Ability to OCR an image with printed text and provide the output as plain text or HTML.
2. Ability to OCR an image with handwritten text and provide the output as plain text or HTML (this is less of a priority).
3. Ability to OCR an image and return any barcodes contained therein (somewhat less of a priority).
4. High degree of accuracy of text extraction is key.
5. Support for various languages is required: English and various European languages, Chinese/Japanese/Korean, Arabic, Hebrew.
6. The operating systems are Windows and Linux.
7. We would want a Java API for this.
8. The image source may be arbitrary (scanners, etc.)

Thanks.

7

Hello,

Thanks for considering Aspose.

We will put resource to investigate about the solution. I will update you if we get any progress. However, OCR is a different story to Barcode recognition. I cannot expect we will have a good solution in the short term.

Best regards.

8

Certainly understandable. I'd really like to get a general sense of when this capability could become available. E.g. Q3 or Q4 of this year or is this an even longer process.

What's clear at this point is I see a lot of interest, from Web posts, in Java-based OCR. Unfortunately it looks like none of the existing Java-based OCR offerings deliver or live up to the expectations: their quality is typically very poor. Therein lies a good business opportunity for a robust Java-based OCR offering (can be pure Java or, say, a JNI wrapper of C-based functionality).

Thanks.

9

Hi dgoldenberg,

I have merged your new thread with an existing one because both are quite relevant.

Absolutely we're quite interested in making Aspose.OCR for .NET and Java. I'm quite busy with building our business development teams these days. Once these teams are established I will switch back to build product development teams. OCR is a large and complicated area so we do need to a new competitive product development team dedicated to that.

I thank you for your patience. I will keep you posted when we have made progress.

10

We are planning to release a new product called Aspose.Recognition soon. At the moment it will support PDF to any output format supported by Aspose.Words (DOC, RTF, OOXML, ODF, HTML, TXT).

Converting PDF to these formats is really a "recognition", not a simple conversion, that's why it is a separate product. If the product lives up to the expectations we will add an OCR module to it too.

Also we are trialling a technology for automatically porting our .NET solutions to Java and if that goes well you can expect to get this product for Java too.

I cannot promise an OCR for Java this year, but stay tuned, things might be going in that direction.

11

You can help us if you zip and attach some of the images with text that you want to recognize.

12

Hi all, I have close to the problem raised above. Am able to read and .tif image but am not sure what to do to read the hand written characters into text. Anyone with an idea or some notes? Am using JAI of java.

Mutuah

13

Hi all, I have close to the problem raised above. Am able to read and .tif image but am not sure what to do to read the hand written characters into text. Anyone with an idea or some notes?

Mutuah

14

Hello,


We found this topic very interesting.

We are looking for a PDF OCR to convert PDF with pictures inside to text recognised PDF. As said in this topic, it seems that ASPOSE was working on it in 2008.

Did you release these developpements on PDF recognition. If not, are you still considering these developpements ?

Thank you in advance,
Laurent VAILLANT

PACIFICA
00 33 (0) 1 53 74 33 66
15

Dear Laurent,


We would like to let you know that we have started working on OCR server component, which is planned to be integrated in our Aspose.Pdf product line in due time.

It is too early to specify exact dates, but we plan initial OCR component functionality to be available for public review in Q1 of 2011.
16
Do you have any updates on this ?
17

Dear Ujjwal,


I can confirm that we are working on OCR component (currently it’s not in state for public release, but hopefully first version of it will be available to work with Aspose.PDF for text recognition in either April or early May).

Should it happen earlier, we will keep you posted.
18

Does this component exist today for the .NET Framework?

19

Hi Vitaly,


I’m wondering if the feature is already available as Aspose.Pdf.Kit?
Or is it only extract text in PDF, instead of image of a text in PDF?

Regards,
Dodi Darundriyo
20

Hi,

Any news about this topic?
We need an ocr component, and would like to use the one from aspose if there is one…

cheers
manuel