17-03-21 04:45 PM
18-03-21 09:29 AM
19-03-21 04:58 PM
22-03-21 10:29 AM
03-07-23 07:19 PM
Did you find an answer for this Tejaskumar? We have a weird scenario where decipher identifies a comma before a number which isn't visible to human eye. Once we refresh region in data verification stage, it disappears.
04-07-23 05:51 AM
To my understanding, the "Refresh Region" right-click option re-extracts the value of that specific box.
04-07-23 05:53 AM
Venkata Pranav Reddy Thathireddygari If possible can you add a snap of that field where you face this issue? Also what happens when you redraw a new region?
05-07-23 05:43 PM
The region around the number 366 is being taken as ,366 (with an added comma). Once we refresh the region around 366, it disappears and correct number is picked (without the comma).
PS: Please ignore the quality of screenshots. This is all I could get immediately.
06-07-23 08:02 AM
Hi All,
The refresh regions button will refresh the identified region(s) using the "on-demand" OCR engine, which is different from the one used during the main OCR stage. This is in part because Decipher also extracts data from PDFs programmatically, also the primary OCR engine is not suitable for use during the verification stage. As such you may get different results, especially where image quality is not optimal.
What data type is the field set to? If it's not integer, that may help resolve the situation. Alternatively using an "STRREPLACE" formula can be used to remove the character or a format expression e.g. [0-9]{1,5}
Thanks
Ben