We are continuing to identify ways to ensure our Invoice Validation Interface allows users to extract the data from invoices with 100% accuracy. To that effect, we have launched two new functionality improvements that will allow users to better deal with some exceptions that come up from time to time.
To reduce processing times and to ensure accuracy, Affinda only applies Optical Character Recognition (OCR) technology to documents that are not machine-readable. When a document is uploaded, Affinda will check to see if this document has a 'text layer' associated with it. If it does, then no OCR is applied as it assumes that this text layer matches the document.
However, occasionally the text layer will match the document correctly for various reasons and will mean that the user cannot validate the data correctly (other than to re-type each data field in our tool).
To address this, we have added the ability for users to manually choose to run OCR on a document to get better text accuracy. To do so, simply click the vertical ellipsis at the top right of the document and select 'Run OCR'.
Previously, users could only select text from an invoice using a single rectangle. In most cases within an invoice, a single rectangle will select all the required data. However, occasionally a user would need to select text using multiple rectangles or when a single large rectangle would capture additional text not required.
The Validation Interface now supports multiple rectangles per annotation. To do this, while an annotation is selected, hold down 'Control' to draw additional boxes. You can select individual boxes to resize them and delete them with the Delete key.