
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content
<(RM) Support tip> How OCR rendition requests are submitted and processed
Incident description :
This Support tip post explains how OCR rendition requests are submitted and processed.
Resolution :
1, Pre-requisites
IDOL Image Server and Key View are required function module for OCR requests.
2, When submit an OCR request for a record.
When request for OCR rendition is submitted by user, HPE CM will check the render queue table TSRENDQUEU ( queued job ) and TSRENDFAIL (failed job) for OCR request for this record , if there are no existing queue or failed job and the OCR rendition doesn’t already exists ( TSRECORD table rcHasOCRFmt = F ) , a request will be added to the TSRENDQUEU ( rendition type = 1 for OCR rendition) , and Render will process the requests from there.
3, Workgroup server assigns the request to the Render server, and Render extracts the documents from the document store to the IDOL Staging directory.
4, Render hands over the request to Image Server.
Image Server uses Key View to extract the text from the image block or document, and write the text to an OCRX file under IDOL Staging directory.
5, Once Render finish writing text to OCRX file, the request will be handed over back to Render, and Render will check in the document to HPE CM and attach the OCRX file to the record as OCR rendition.
6, Render clears the IDOL Staging file directory.