Enable table extraction: Select this option to extract data from tables, sometimes called spreadsheets, which are typically defined by columns and rows in a grid format.

• Minimum spaces between works to signal column breaks (2-10): Defines the table spacing of the form to fine-tune the OCR recognition of the region. The number of spaces serves as a clear delineation to identify the break between columns and rows based on the standardized spaces between characters.

• Fail classification if table extraction does not produce any records: Ensures that table data must be extracted for the form to pass classification form matching. Use this setting to ensure that the type of form being identified must have table data, which is then extracted into records in order to be a successful match.

  Specify a search type to determine how PSIcapture decides whether a classification form fails to match if table extraction does not produce any records.

• Search type: Specifies the search type for table extraction from the following options:

  • Stop search on first non-matching line: As soon as a non-matching line is recognized via OCR, the search stops and PSIcapture does not use any more processing power and marks the classification form as non-matched.

  • Search to bottom of page: The search is more in-depth to ensure there is no table data to be extracted from the entire page. Using the same logic as the other option, PSIcapture continues the search for the entire page until it determines that table extraction is not possible.

• Preview: Previews the table extraction settings on the current classification form image.