Transym OCR 4

Enable this option to automatically detect and invert regions of the image that contain white text on a black background.

• Recommended for documents that may contain negative (reversed) print zones.
• Helps improve recognition accuracy for forms, checks, or documents with mixed print styles.
• Leave enabled unless you observe unwanted inversion on normal text.

Deskewing corrects images that are slightly rotated or slanted, improving OCR accuracy.

• Recommended for scanned documents where pages may not be perfectly aligned.
• Disable only if you want to preserve the original skew or if deskewing causes issues with your layout.

Deshading removes background shading or gradients that can interfere with text recognition.

• Useful for forms, checks, or documents with shaded areas or colored backgrounds.
• Leave enabled for most documents unless you observe loss of important background features.

Inverts all colors on the page, turning black-on-white to white-on-black and vice versa.

• Use only if your entire document is printed in reverse (white text on black background).
• Not typically needed for standard documents.

Removes small specks, dots, or other noise from the image to improve recognition.

• Recommended for low-quality scans or documents with background artifacts.
• Disabling may be appropriate for very clean, high-resolution images.

Removes horizontal and vertical lines (such as table borders or form lines) that may interfere with text recognition.

• Useful for forms, tables, or documents with grid lines.
• Disable if you need to preserve lines for downstream processing or if text is being removed unintentionally.

Can be one of the following values:

Enables detection of document sections, such as headers, footers, or columns.

• Useful for complex documents where logical structure is important.
• May increase processing time; disable for simple, single-column documents.

Attempts to join fragmented or broken characters into single, complete characters before recognition.

• Recommended for low-quality scans or dot-matrix printouts where characters may be split.
• Leave enabled unless you observe incorrect merging of distinct characters.

Filters out lines that are likely to be noise, stray marks, or non-text artifacts.

• Improves output quality for noisy or degraded documents.
• May remove valid but low-confidence lines in rare cases.

Removes characters that the engine determines are likely to be incorrect or noise.

• Use to improve accuracy on documents with significant background noise or artifacts.
• May remove valid but low-confidence characters in some cases.

Uses a built-in dictionary to improve recognition of valid English words and correct common OCR errors.

• Recommended for most documents containing standard English text.
• Disable if your documents contain codes, part numbers, or other non-dictionary content.

Restricts recognition to a specific set of characters, improving accuracy for forms, codes, or documents with limited character sets.

• By default, includes all printable ASCII characters.
• Edit this property to limit recognition (e.g., to digits only for numeric forms).
• Leave as default for general English text.