System and method for detection and removal of thin lines in a graphic image

A method for identifying and removing one or more thin lines from a binary source image, the method comprising the steps of:

• (a) generating a hole-filled intermediate binary image, by performing the sub-steps of identifying holes applying a Connected Component Labeling, Two-pass Algorithm to the binary source image;
• (ii) filling the holes identified in step (a)(i);
• (b) performing a thin line detection and removal process on the hole-filled intermediate binary image, by performing the steps of:
• creating an eroded intermediate image by performing a morphology erosion process on the hole-filled intermediate binary image using a first kernel size;
• (ii) creating an opened intermediate image by performing a morphology dilation process on the eroded intermediate image using a second kernel size larger than the first kernel size;
• (iii) inverting the opened intermediate image resulting from step (b)(ii);
• (iv) generating a density adjusted binary source image by filtering the binary source image using a user-specified maximum density threshold, so that the density adjusted binary source image file includes only pixels corresponding to tones below the maximum density threshold;
• (v) generating a difference pixel intermediate image by performing a binary AND operation between the result of steps (b)(iii) and (b)(iv), the result of which represents pixels and thin lines to be removed from the source image;
• (vi) generating a thin line subtraction intermediate image by identifying and removing scattered dots from the difference pixel intermediate image by applying a Connected Component Labeling, Two-pass Algorithm to the difference pixel intermediate image;
• (vii) subtracting the thin line subtraction intermediate image from the binary source image to create a binary output image.