Despeckle is the process of removing speckles from images (especially bitmaps created using a scanner). Speckles are artifacts which are extra pixels (or collections of extra pixels) that can occur in scanned images because of imperfections in the scanner hardware, or because of dust or dirt being present on the camera, scanning surface or even the document being scanned.
The document structure analysis and character recognition are usually done in several phases:
scanning
thresholding
skew detection and correction
despeckle or speckle removal
line removal
border removal
detection of preprinted elements (like boxes)
page orientation detection and correction
layout analysis
classification
character recognition
Each step must be completed well enough for the performance of the sequence and result to be successful. Steps that follow the despeckle are inefficient if the correction fails.
DespeckleHelper provides a successful method for automatically improving the quality of document images and a marked decrease in OCR error rates. The method is easy to use - we view it as a pre-OCR cleanup operation, and it takes a very small amount of the computational effort of the OCR process itself.
DespeckleHelper expects as input a monochrome image.
