Handwritten forms with image artifacts

In a lot of organizations and particularly in the federal, state or city administrations, most data processes start with the capture of information from forms. Forms have several types of complexities including handwritten entries, image artifacts such as selection marks and circled text, and tabular data.

iQC learns from the domain experts to capture all of these types of data with the same and uniform point-and-click user interface. The  capture of hand-filled fields is based at on the detection of text  and the recognition of the context where it occurs. It is also easy to train our AI to capture tabular data and localized image objects such as circles around text and selection marks in any types of forms .

shutterstock_67555918.jpg

We have trained iQC to recognize these different types of data from complex forms for the US Air Force. iQC learns to read form data like a human by focusing on the context of each data entry rather than learning rigid form templates. This makes iQC robust to changes in form layouts and versions. iQC also allows the user to enter domain specific information to make the form reading easier and to capture data in spite of OCR errors.