Skip to main content
Machine-readable forms are designed for automated data input, i.e. the program should be able to easily distinguish between the data to be captured and the non-recognizable form elements, such as field borders, text marking, backgrounds, or explanatory text. This can be achieved in one of two ways:
  1. The form must be designed in such a way as to make the field borders, background and text marking disappear after scanning, or
  2. The form must be designed in such a way as to make the field borders, background and text marking stand out clearly after scanning.
If the first approach is used, only the anchors, explanatory text, and entered data will remain on the form after scanning. As a result, the program will not have to separate the data from form elements, which will make recognition more easy and reliable. In this case, the following types of forms are used: dropout forms, black-and-white forms with raster backgrounds, black-and-white forms with raster borders, and black-and-white linear forms. If the second approach is used, the anchors, explanatory text, entered data, as well as text marking and field borders will remain on the form after scanning. The program will look for information that is surrounded by field borders and text marking. If these form elements are missing, recognition quality will be adversely affected. In this case, black-and-white linear forms are used.