You are viewing limited content. For full access, please sign in.

Question

Question

Automated Quality Assurance Process

asked on April 1, 2016

With the volume of items imported (scanned to network drive first then imported into Laserfiche) it may not be possible for human intervention to confirm that these items were scanned correctly.  We would like to dispose of the paper document, but want to confirm the documents scanned correctly before doing so without having to physically view each one. 

Is there any type of automated quality assurance process to determine if a document scanned correctly?

Thank you.

 

0 0

Replies

replied on April 1, 2016

Hi, 

I'm a bit confused about exactly what you're looking to do. Can you give examples of some issues you'd want to be able to detect automatically?

0 0
replied on April 4, 2016

Tessa,

Specifically is there any way within the Laserfiche Suite or by using another piece of software that we could identify the following situations:

image quality - is quality of image readable?  Not sure if this could be some measure of the success of OCRing the document or not.

 

content of image - did the entire page scan? There were no folded sheets or double feeds. 

 

there are pages in the document - sometimes you get the document placeholder but no page images.

 

template completed

Basically all the things that a scanner operator would be evaluating for.  The scenario is that the user will scan to a network drive and from there Laserfiche will import and Workflow will route the document.

 

Looking  for thoughts on this possibility.

Michelle

0 0
replied on April 4, 2016 Show version history

Hi, 

I want to start by addressing the last two you mentioned (checking if the template fields are filled out and checking if there are pages in the document). These two should be pretty straightforward using Workflow or Quick Fields or a combination of both. If you're just looking for an automated way to find documents without pages or metadata so you can make corrections by hand, then a simple Workflow can do this. However, if you also want to make the corrections with minimal or no human intervention, then you'll likely want to use Quick Fields. It will of course depend on exactly what data you're trying to extract, but for most cases, Workflow and/or Quick Fields will have the tools to do it. 

 

Now to get back to the first two issues you mentioned: 'Image Quality' and 'Content of Image'. These are the trickier ones here because unless there's a specific criteria for success/failure, these things are really subjective and thus may require a human eye. However, you could make the manual workload much more manageable by automating some of the more straightforward cases. For example, you could use Quick Fields to identify pages that are seem to be blank or are missing text in a certain area, and then scanner operators would only have to manually inspect those 'problem' pages/documents rather than everything. 

0 0
replied on April 5, 2016

Tessa,

The scenario for the last two items (paragraph 1) is that we would like to make the corrections with minimal or no human intervention.  Regarding the items in the second paragraph you correctly identified what we'd like to accomplish.

Big question is how do we do this?  Are there any recommendations or reference information you could share on how this would look? 

Thanks!

Michelle

 

0 0
replied on April 5, 2016 Show version history

Sure, https://www.laserfiche.com/feature/capture/, especially the "Accelerate high-volume capture" section, would be a good place to start. Note that Quick Fields processing can be run on documents already in Laserfiche or on a network drive in the same way they can be run 'at scan time'. 

1 0
You are not allowed to follow up in this post.

Sign in to reply to this post.