Line Removal reduces OCR errors, especially in images where text and lines are placed closely together (in underlined text, for example). You can activate Horizontal Line Management and Vertical Line Management separately with independent parameters. Other than orientation, these features are identical.
|
Option |
Description |
|---|---|
|
Max. Character Repair Size |
Automatically reconstructs intersected characters after performing line removal. This parameter sets the maximum width and height, in pixels, of characters to reconstruct. Increase this value if text is not adequately repaired, and decrease it if you encounter erroneous reconstruction. Text larger than 14 points may require higher settings. Setting this value to zero disables character reconstruction. Range is 0-100. |
|
Max. Gap |
Lines in a scanned document often contain small gaps. With Max. Gap, you can set the maximum gap (in pixels) to be considered a continuous line. Activating this feature reattaches broken segments for removal purposes. For poor quality images, such as dot matrix and microfilmed documents, set this value as high as 20. With high quality scans, set the Maximum line gap to 0. Range is 0-20. |
|
Max. Thickness |
Sets the maximum thickness, in pixels, a line can have to meet the removal criteria. Since large text is usually thicker than the lines on forms, use the Max. Thickness feature to keep very large text in titles from being removed. Range is 1-50. |
|
Min. Aspect Ratio |
Sets the ratio of the line length to the line width. Range is 1.0-1000.0. |
|
Min. Line Length |
The line removal system can identify and remove very short lines, such as the one across a large capital T. Min. Line Length sets the minimum length of a line (in pixels) to remove. Set this value larger than the height and width of the text characters. Range is 10-20,000. |