Showing results for 
Show  only  | Search instead for 
Did you mean: 

The extracted data IRIS returns from images is organized according to the following different IRIS modules:


Document Image Classification (DIC):

Returns the document type (“docType”) based on categories used to train the model; the level of confidence ("confidence") ranges between [0, 1.0], where higher confidence values (i.e., close to 1.0) mean the greater the chance of prediction being right.

Optical Character Recognition (OCR)

Returns a list of texts extracted from the image, where the list is organized by paragraphs, i.e., each item of the list contains the text corresponding to a paragraph. Additionally, the list order follows the top-down and right-left patterns.

Document Field Classification (DFC)

Returns a dictionary, where each key is a field name (e.g., date, zip code, ...) and a value is a list containing all field values that are matched with the pattern configured.


You can use the data extracted by IRIS for different reasons. For example, you can display the extracted information to the end-user and ask to them to confirm it, you can use the information to update your database, and so on.


Important: When an image is classified as fake, IRIS interrupts its flow and returns a notice that the image is fake without executing the OCR and DFC features.


To learn more or to help you get started with IRIS, feel free to contact Sinch or if you are an existing customer reach out to your Sinch Account Manger.

Version history
Last update:
‎08-30-2022 07:45 AM
Updated by: