From this article you will learn:
- How to split documents using separator page (for scanned documents)
- How to enable split suggestions for uploaded documents
- How to manually split documents
- How to set up advanced splitting logic
- What are the current limitations of splitting
How to split documents using separator page
Rossum provides several ways to split documents, and one of them is our separator page. This feature is helpful when you have multiple documents stored in one file to scan and upload to Rossum. To use it, you simply print a special page with a QR code on separate pieces of paper and insert these pages between the documents while scanning. When you upload the file to Rossum, the system will quickly recognise the separator pages and create separate entries for each scanned document in your queue.
How to enable split suggestions in Rossum
Note that the split suggestion feature is not available for queues using the Dedicated AI Engine.
If your documents are directly uploaded to Rossum (e.g., sent via email by vendors or clients), you can use split suggestions, which can be configured on a queue level. To enable this feature, follow the steps described below.
First, open your queue settings.
In “Basic settings,” scroll down to the “Split batch files” option, select “Suggest” from the dropdown list and save the changes.
Once enabled, Rossum will try to identify batch files uploaded to the queue and suggest the split. Documents with split suggestions will be marked with an orange scissors icon on the annotation list.
Clicking on the scissors will take you to the “Edit document” screen, where you can proceed according to the instructions provided here.
By default, Rossum’s AI Engine captures data only from the first 32 pages. The same limit applies to the detection of possible splits. If needed, this limit can be extended. Do not hesitate to contact us at firstname.lastname@example.org if you want to increase this limit.
How to manually split documents
You can manually split any document that has more than one page in Rossum.
Open the document you want to split and click on the scissors icon on the right side of the annotation screen.
You will be directed to the “Edit document” screen, where you can split the document by simply clicking between two pages. For further details, check out this link.
After the document is split, new annotations (children) will be created, and the original one (parent) will no longer be available in the queue dashboard. You can check the original file and adjust the split later, as described here.
How to set up more advanced splitting logic
For more complex splitting requirements, Rossum offers the Document Splitting Extension. With this extension, you can create configurable rules for automating document splitting. Some examples of rules are:
- Fixed number of pages
- Presence of specific text
- Splitting based on the value of a field extracted from a document
The Document Splitting Extension enables reliable automation of document splitting, even for queues that use the Dedicated AI Engine and cannot benefit from the default split suggestion feature.
What are the current limitations of splitting
Currently, Rossum does not support splitting the following file types:
- .doc and .docx
- .xls and .xlsx