|
Form
Processing
Forms can be captured by two ways - Automated and
Manual.
Automated Form
Processing
-
Data 'extraction' is
done from Scanned Digital formats of forms making
use of software. Our unique zone-specific image
processing techniques have few advantages over
others. Zone-specific image processing is more
focused as it impacts only a specific extraction
zone.
These zones are defined in our unique Form
Development Environment where each extraction zone
is assigned a 'text area' or 'text control' on a GUI
Form. This approach has other benefits apart from
faster results - before saving the OCR data to a
database etc., the entry can easily be checked for
correctness; some features such as de-shading,
de-skewing, character repair etc., need only be
applied to these zones rather than full page thus
increasing the processing speed and thus enabling
large volume production. Our Data Verification
Process ensures high degree of accuracy. Where the
required accuracy level tends to 100%, we set for
blind verification by two separate verifiers, with
their respective decision compared and accessed by a
supervisor.
Manual Form
Processing -
Data
'extraction' of forms is done by transcribing
manually. (Single or Double Data Entry) Automated
extraction via a recognition engine usually fails
for certain types of data, including cursive
handwriting as an example. Our Form Development
Environment provides access to many proprietary
ActiveX controls (Custom controls implementing
features like auto-completion, desired date format,
spell-checking option etc., with/ without validation
rules) to facilitate fast data entry. Form
Development Environment also integrates well with a
range of known Databases including MDB, DBF, MS
Excel and sequential text file output. Our XML
editor integrates with these databases to create the
desired XML output when required. As an extended
service, we also carry out archiving, data analysis
and reporting. |
