LEADTOOLS OCR Send comments on this topic. | Back to Introduction - All Topics | Help Version 17.0.3.22
Cleaning Up 1-Bit Images

(Document/Medical) Scanned documents often contain a number of undesirable elements, including dots, blobs, lines, borders and hole punches. LEADTOOLS provides a number of classes that will improve the appearance of scanned documents by removing dots, blobs, lines, borders, hole punches, inverted page, inverted text, bumps and nicks. These classes are listed in the following table:

Leadtools.ImageProcessing.Core Class WCF Support WF Support Description
SmoothCommand Smooth

SmoothActivity

Smooths the bumps and fills in the nicks of a 1-bit black and white image.
BorderRemoveCommand BorderRemove

BorderRemoveActivity

Removes the black borders in a 1-bit black and white image.

LineRemoveCommand

LineRemove

LineRemoveActivity

Removes horizontal and vertical lines in a 1-bit black and white image.

DotRemoveCommand

DotRemove

DotRemoveActivity

Finds and removes dots and specks of various sizes.

HolePunchRemoveCommand

HolePunchRemove

HolePunchRemoveActivity

Finds and removes hole punches.
InvertedTextCommand InvertedText

InvertedTextActivity

Finds and modifies areas of inverted text in a 1-bit black and white image.
InvertedPageCommand InvertedPage

InvertedPageActivity

Checks and auto-correct an inverted image (1-bit or color).
HighQualityRotateCommand HighQualityRotate

HighQualityRotateActivity

Perform high quality rotation on a 1-bit black and white image.

Each of these classes utilizes flags that allow you to obtain a LEAD region (Leadtools.ImageProcessing.Core only) or a Windows region (all three) that contains all the elements removed from a document. In addition, event handlers allow you to provide customized region processing, as well as create your own composite of the removed regions.

Images can be straightened by using the following commands:

Leadtools.ImageProcessing.Core Class WCF Support WF Support Description
Deskewing.

For information about pre-processing color images, see Pre-processing Images for OCR.