Tesseract OCR

Can be one of the following values:

Can be one of the following values:

Can be one of the following types:

Select the language for text recognition. Additional languages can be downloaded from the Tesseract Data Files page.

• Place downloaded language files in the Tesseract\\tessdata folder within your Grooper installation.
• Not all languages are supported by Grooper (e.g., vertical or right-to-left scripts may not be recognized).
• Changing this property will also update the 'Character White List' to match the selected language's character set.

Restrict recognition to a specific set of characters to improve accuracy, especially for documents with limited or known character sets (such as numeric forms or codes).

• This value is initialized automatically when the 'Language' property is set, but can be edited manually.
• Leave this property empty to allow all characters supported by the selected language.
• Use to reduce false positives and improve recognition speed on structured documents.

Enable recognition for specialized fonts by adding them to this list. Grooper ships with pre-defined support for:

• MATH: Math symbols and equations.
• MICR: E13B MICR font for US checks (characters: 0123456789ABCD).
• OCRA: OCR-A font (Basic Latin character set).
• OCRB: OCR-B font (Basic Latin character set).

Add the desired font(s) to improve recognition accuracy for documents using these styles.

Controls how wide a gap must be to qualify as a space character. The "kern size" is the average small gap between adjacent characters.

• Increase this value if false spaces are being detected.
• Decrease this value if valid spaces are being missed.
• This property corresponds to the Tesseract parameter tosp_min_sane_kn_sp.

Typical values range from 1.0 (100%) to 2.0 (200%).

By default, Tesseract will detect and read vertical text segments (such as text printed along the page margin).

• Enable this option to remove vertical text from the OCR output, which is useful for forms or documents where vertical text is not relevant.
• Disabling this option allows all detected text, including vertical segments, to be included in the results.

Use this option to filter out words that Tesseract identifies as likely to be noise or artifacts, improving the quality of the recognized text.

• Recommended for documents with significant background noise or stray marks.
• May remove valid but low-confidence words in some cases.

Use this option to filter out lines that Tesseract identifies as likely to be noise or artifacts.

• Useful for removing spurious lines from noisy or degraded images.
• May remove valid but low-confidence lines in some cases.