12 likes
·
1.7K reads
6 comments
Thanks for sharing this Jon! Actually, I was looking at this OCI service today for a project. Did you try the TABLE_DETECTION option?
I haven't, but I will also be working on this more for a project I am working on.
Thank you for providing all relevant links for this service, including the Git docs - appreciate the comprehensiveness of your posts. Thanks Jon!
Glad you liked it.
Hey Jon, Thanks for such interesting content, now I have a question for you: In this example you are first scanning the documents and from there you are uploading them to Oracle Storage, and from there the documents are analyzed with document AI, perfect.
But this is where my question comes from and I have not found an answer in any article or blog, and it is how to connect peripherals to APEX. In a specific case, connect a Check Scanner to the computer and browser where APEX is running. Tasks to perform from APEX: Open the Check Scanner in APEX (screen), and be able to capture the image in real time, and from there you can follow the process of saving it in a Bucket, analyze it with (OCR) and from there be able to extract the data: such as the bank account, value of the check, etc, and to be able to put them on the screen, to be reviewed by the user and if there is any change do it (the OCR is not always perfect) and from there to be able to save the data in the Database.
Could you help me with this, and it could also be a good article, how to connect peripherals to APEX
Hi Mike. Good question, connecting peripherals directly using a Web App is a challenge. In previous projects, I have utilized OneDrive. 1) Users scan to a local directory which is Synced to OneDrive using OneDrive App. 2) User pulls up a scan in APEX using MS Graph APIs to fetch files from OneDrive (similar to the approach in this blog. 3) Process the file based on button click in APEX.
Variations on this include polling the OneDrive directory and posting files to Object Store then using Object Store Events to process the file. Similar to this blog.
Finally, you could have a local script on the user's PC that posts scanned files directly to Object Store using an OCI Pre-Authenticated request (blog post on this coming soon).