Automated Extraction


[shortmenu menu=”Square 9″  display=”inline”  enhance=”true”  submenu_color=”#000000″  submenu_anchor_color=”#e0e0e0″  submenu_anchor_hover_color=”#ffffff”  submenu_transparency=”0.8″  arrow=”true”  is_responsive=”true” ]


Automated Extraction


Optical character Recognition

Your paper based documents are a source for business critical information. Most commonly this information is reviewed and then entered manually into your line of business applications where it’s capture can be prone to human error. At Square 9 we believe that reliably extracting high value information where it can be shared and re-purposed is one of the critical benefits a document management system can provide.

Square 9 delivers advanced data extraction with Optical Character Recognition (OCR)tools that will change the way you look at capture automation. Powerful image enhancement options, page registration for the correction of paper shifts during the scanning process and confidence based reporting are just a few of the tools that combine to ensure that you are getting the highest quality data extraction from your documents.

We then feed the resulting data through a Quality Assurance engine that checks your output prior to releasing it to SmartSearch. Questionable results that fail to meet the assigned level of confidence are re-directed to a Batch Manager for further review. Any reported irregularities can be quickly audited from the Batch Manager and corrected on the fly. This includes not only OCR quality issues, but also errors that can occur such as character mismatches or data truncation which might impact your output. The final result is quality data you can count on.

Incorporate Confidence in your OCR

Design your template

With our template designer you can quickly load your sample documents, establish the OCR zones and set the properties for your data. These include data orientation, type and the level of confidence you require from the extraction. To take accuracy to the next level, add a page registration marker to protect against misreads that result from paper shifts during the scanning process.

Add Text Extraction to your Capture Worklow

The Text Recognition Activity is simply dragged into your pre-defined document capture process where it goes to work grabbing data. That means you can combine text capture with other activities like Image Enhancement, Automated Page Separation and Full Text PDF Conversion. It also means documents can come from a variety of sources including multifunctional printers (MFP), Desktop Scanners, or even from a watched network directory.

Be Confident in your Output

Behind the scenes our quality assurance engine is examining your OCR process and reporting on the results. Data successfully extracted from your documents will be released to SmartSearch along with the original image file. Questionable results will be re-directed to our batch manager where they can be manually reviewed, edited and released into production once full integrity has been achieved.