OCR Reader

The 'OCR Profile' property specifies which OCR Profile to use when extracting text from the defined region.

• If set, the selected profile's OCR engine and settings will be used to perform OCR on the region.
• If left blank, text will be extracted from the document's existing full-page OCR results.

Use this property to override the default OCR settings for specific extraction scenarios, such as when higher accuracy or different preprocessing is required.

The 'Region' property defines the extraction zone as a logical rectangle, relative to each output instance.

• If left blank, the region of each instance will be used as-is.
• Specify a region to extract text from an area adjacent to or offset from the instance, such as a value to the right of a label.

Use this property to target specific areas of the document for OCR extraction.

Can be one of the following values:

The 'Relative To' property determines the anchor point of the region relative to each output instance.

• Choose the edge or corner (e.g., TopLeft, BottomRight) that the region should be positioned relative to.
• This allows for flexible extraction of values that appear in a consistent position near each instance.

The 'Auto Snap Distance' property enables the auto snap feature, which adjusts the extraction region to align with detected lines (such as table borders or form lines) in the document.

• Set a value to enable auto snap; leave blank to disable.
• The value defines the maximum distance (in inches, points, or percent) to search for a line to snap to.

Use this property to improve extraction accuracy in documents with well-defined lines or zones.

The 'Auto Snap Margin' property allows you to shrink the extraction region by a specified amount on each edge after auto snap is applied.

• Use this to avoid including unwanted borders or whitespace in the OCR region.
• Only applicable when 'Auto Snap Distance' is set.

Can be one of the following types:

The 'Value Extractor' property allows you to specify a Value Extractor to run on the OCR results.

• Use this to extract, validate, or filter specific patterns from the OCR text, such as numbers, dates, or custom formats.
• If not set, the raw OCR output will be used.

This is useful for refining extraction results or enforcing data validation.

The 'Discard Misses' property controls the output when the Value Extractor does not find any matches in the OCR results.

• If set to true, the output instance will be discarded.
• If set to false, the original OCR result will be used as the output.

Use this property to suppress empty or invalid results when strict extraction is required.

The 'Value Separator' property defines how multiple values extracted by the Value Extractor are joined in the output.

• Enter a string or escape sequence to use as the separator.
• Supported escape sequences:
  • \r - Carriage return
  • \n - Line feed
  • \t - Tab
  • \f - Form feed
  • \s - Space

Example: Use \n to join values with a line break, or , for a comma-separated list.

The 'Line Separator' property controls how line breaks in multi-line OCR results are represented in the output.

• Enter a string or escape sequence to use as the separator.
• Supported escape sequences:
  • \r - Carriage return
  • \n - Line feed
  • \t - Tab
  • \f - Form feed
  • \s - Space

Example: Use \s to join lines with a space, or \n for a line break.

The 'Exclude Anchor' property, when set to true, removes the text of the anchor or label from the OCR results.

Use this option to prevent the label or anchor text from appearing in the extracted value, especially when extracting values adjacent to field labels.

The 'Output Full Region' property controls how the highlight region is displayed for each output instance.

• If set to true, the full OCR region (as defined by the extraction zone) will be highlighted.
• If set to false, only the bounding box of the recognized text will be highlighted.

This option is useful for tuning and visualizing the extraction region during configuration.