Currency Decimal recognised as number

Problem -> Decimal are being recognised as the number zero ‘0’
image.png (59.1 KB)

PDF being converted to .jpg using Aspose jpegDevice with resolution 300 and quality 100
image.png (13.9 KB)

Using recognition block to get text, have tried with and without filters
image.png (46.9 KB)

@support-23

Thanks for contacting support.

Would you please provide us the sample image upon which you are performing OCR operation. We will test the scenario in our environment and address it accordingly.

Hi please find the attached the PDF and image, it is the GST amount that is the issue.

Please note i found this same issue with other images as well.Invoice INV-12881 Accounting Scott Partners.jpg (464.0 KB)
Invoice INV-12881 Accounting Scott Partners.pdf (72.4 KB)

@support-23

Could you please share the values of intX, intY, intWidth, intHeight which are used in RecognitionBlock.CreateTextBlock() method so that we can test the scenario accordingly.

They are 2239, 1675, 136, 58
Please also see the imageAspose COORDS.jpg (114.6 KB)

@support-23

We were able to replicate the issue in our environment and logged it as OCR-694 in our issue tracking system. We will further look into details of it and keep you posted with the status of its correction. Please be patient and spare us little time.

We are sorry for the inconvenience.

I’m having more issues in the Money value the comma ‘,’ is being recognised as a decimal point ‘.’
and in the date the forward slash ‘/’ is being recognised as square bracket ‘]’

please see the images for description and PDF and image files

These are very standard things that that don’t seem to work, please advise i may have to get a refund based on that fact it cannot do basic scanning.

BPay Creditor Invoice.jpg (1.2 MB)
BPay Creditor Invoice.pdf (87.7 KB)
Decmial Point problem.jpg (133.8 KB)
Date error.jpg (141.6 KB)

@support-23

We are checking this and will share our feedback with you soon.

Hello, can you please provide some feedback on the issues i need to deliver my product within 2 weeks and will need to use something else if i can’t get a fix by then thanks.

@support-23

Thanks for your patience.

We were able to replicate the issue in our environment while testing the scenario using your images. We have also observed that API gave correct output (i.e. commas were recognized correctly) when we did not use RecognitionBlock. However, the issue remained for slash in date format.

Therefore, we have logged this issue as OCR-706 in our issue tracking system for further investigation. We will look into details of the issue and keep you posted with the status of its correction. Please spare us little time.

We are sorry for the inconvenience.