Preprocessing of dot-matrix/ink-jet printed text for Optical Character Recognition

A method of processing image data of dot-matrix-ink-jet printed text for Optical Character Recognition (OCR) comprising the sequential steps of:

• (a) detecting whether dot-matrix or ink-jet printed text is present in the image data;
• (b) determining the image characteristics of the dot-matrix/ink-jet image data detected in step (a) by forming a histogram of density values of pixels of the image data;
• (c) reducing a dynamic range of the image data by altering the structure of characters in the image data from that of dots to one of strokes using the histogram produced in step (b) and a 2-D spatial averaging operation;
• (d) performing a contrast stretching operation on the altered structure of the characters produced in step (c);
• (e) performing edge enhancement of the characters of the image data produced in step (d);
• and
• (f) binarizing the edge enhanced image data of step (e).

Apparatus for processing image data of dot-matrix-ink-jet printed text for Optical Character Recognition (OCR) comprising:

• pre-processing means for pre-processing the image data of the dot-matrix or ink-jet printed text comprising;
• (a) determining means for determining the image characteristics of the dot-matrix/ink-jet image data by forming a histogram of density values of pixels of the image data;
• (b) 2-D spatial averaging means for reducing the dynamic range of the image data by altering the structure of characters in the image data from that of dots to one of strokes;
• (c) contrast stretching means for performing a contrast stretching operation on the altered structure of the characters produced by the 2-D spatial averaging means;
• and
• (d) edge enhancement means for performing edge enhancement of the characters of the image data produced by the contrast stretching means;
• and
• binarizing means for binarizing the edge enhanced image data from the pre-processing means.