Skip to main content
Questionnaires, social security forms, polling slips, and warranty cards are all different types of form used to collect different types of information. Forms differ from other types of document in two ways:
  1. A form always contains a set number of fields.
  2. Each field may contain only a certain type of information. For example, a “Last Name” field contains only last names (if completed correctly) and a “Date” field contains only dates.
Forms are used when information needs to be gathered from a large number of respondents. Manual information gathering is a long and tiresome process, and typos and errors are almost inevitable. Machine-readable forms are used to automate this process. Automated forms processing consists of the following stages:
  1. Setting up the form-processing application (creating a Document Definition and specifying the fields to be recognized).
  2. Acquiring form images (scanning).
  3. Processing the form images (recognizing the images and validating the extracted data).
  4. Exporting the extracted data to an external information system.
Automated processing of forms is most effective for forms that satisfy the requirements described in the following topics.

Machine-readable forms

What makes a form machine-readable.

Elements of machine-readable forms

The elements a form can contain.

Form completion methods

How respondents fill out forms.

Types of machine-readable form

Dropout, raster, and linear form types.

Choose a form type

Compare the form types to pick the best one.

General requirements for machine-readable forms

Design rules for reliable recognition.