Subscribe to Impira's blog
Stay up to date with all things Impira, automation, document processing, and industry best practices.
Creating a holistic automation solution
We pride ourselves on building new features and functionality that actually reduce the time it takes you to process documents. No fancy fluff here — just pure, usable feature richness for your processing pleasure. Which is why we’re excited to share our brand new table extraction functionality and file pre-processing toolkit.
Table extraction pulls key data from tables, recurring lists, and repeating data in docs like invoices, purchase orders, and loss run reports, to name a few. Other OCR solutions claim to do the same thing, but we’ve been hard at work making sure that our table extraction functionality works against even the most complex, unique of tables by being flexible and adjustable.
We had a fantastic group of beta testers who put table extraction through its paces. One particularly fun use case came from Gavin Bridge, Senior Media Analyst at Variety’s Intelligence Platform.
“The data I was working with was locked in PNG format with a lot of information in each row, and I was facing having to manually transcribe it all into Excel. Impira’s tool saved me hours of typing, freeing me up to conduct more analysis.” Gavin Bridge, Senior Media Analyst at Variety
Page rotation, splitting, and deleting — all within Impira
In addition to the table extraction release, we’ve got a few tricks up our sleeves to make processing your docs even easier. You can now do things like rotate or delete a page within a document or split pages of a document into separate files before you start the data extraction process within Impira. This means you don’t have to use another tool to try and do this before you upload a file to the platform.
We’re breaking down some of the specifics below, but if you want to check them out in real-time, you can sign up for a free account and take it all for a test drive.
How do I get started with Impira’s table extraction?
Jump to “How to extract data from tables” for a detailed walk-through.
After you upload and organize similar documents into collections, you can open a file and get set up by highlighting your first row (minus any headers or column names).
Next, label each column header and highlight the corresponding value.
Impira will immediately start piecing together what the structure of your table looks like, and will begin to seek out the same data from other rows.
Impira will also iteratively retrain itself as you continue labeling out the first full row and confirm predictions in other rows. Every time review new rows, Impira will learn, grow in confidence, and retrain itself to be more accurate.
Modify your files: Impira’s new pre-processing toolkit
In an ideal world, we’d all receive files from clients and other departments in perfect order — pages are right side up, split into correct groupings, and contain only the relevant pages you need. We might get there one day, but in the meantime, we have a set of file modification tools to set those wonky things right.
After you upload your files into Impira, you can rotate flipped pages, delete unnecessary fluff, and split pages into separate files before starting data extraction.
Jump to “How to modify files: Page rotation, splitting, removal" for a detailed walk-through.