An Efficient Skewed Line Segmentation Technique for Cursive Script OCR

Algorithm 1

 Preprocessing.

	Input: Grayscale image or RGB.
	Output: Clean image.
	//Begin
	Step 1. //If the image is RGB, convert it into YCbCr color space.
	YCbCr ← RGB
	Step 2. //Eliminate CBCR. And the image becomes grayscale.
	Step 3. //Apply adaptive thresholding.
	Step 4. Apply the global image threshold using Otsu’s method.
	Step 5. Adaptive (mean\median) filter to highlight image features.
	Step 6. Then, apply the Otsu threshold to segment and generate a binary image.
	//End