I'm trying to use Aspose.OCR to read the check number from the upper-right
corner of a check image and from the MICR line. But, I can't figure out what settings to
use. For every set of settings I've tried, it always reads the number
as "EsT? I" for the upper corner. The actual check number is 5317. It always misreads the MICR line to. I would like to have it read the check number in both places and then see if they match. If they do, then I would know that it read it correctly. Otherwise, I could have the user enter it manually.
I tried to upload a cropped image that shows just the part I was trying to process. But, I do not have permissions to upload. I imagine that the font is the same on pretty much all checks. Please advise on what OCR settings I should use or if this is possible at all.
Thanks,
Mike
Hi Mike,
Aspose.OCR has some issues in recognizing the small font sizes and this issue has been logged into our issue tracking system as OCR-29048. We will keep you updated on this issue in this thread.
Regarding attachment issue, I have forwarded this issue to our web development team and they will check this issue. Once attachment issue is resolved, please attach your sample image for testing purpose. We are sorry for the inconvenience.
Best Regards,
Hi Mike,
I hope you are well.
We have fixed the issue with permissions and you should be able to attach the files now. Please try to attach the image now and let us know if there is any issue. We apologize for inconvenience.
I've uploaded a sample of the image I am trying to read. The characters are actually pretty big. I've used other OCR packages that have read characters a bit smaller than this.
thanks,
Mike
Hi Mike,
Currently, Aspose.OCR produces best results if the font size is 32pt +/- 20%. Recognition of smaller font sizes will be improved in the coming releases. We are 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.