On the Performance Improvement of Devanagari Handwritten Character Recognition

Table 2

Proposed feature extraction algorithm.

Input: image (training/test)	Output: features of image
Preprocessing steps	(1) Convert the image into two-tone image; (2) Normalize the image to standard size; (3) Obtain the boundary image; (4) Obtain the features using algorithms 1 or 2 or 3
Algorithm 1 (standard zone)	For boundary image IM(, ); (1) divide the and by number of bins in horizontal and vertical directions to calculate the value of boundary indices and; (2) obtain the subimages generated from IM using and; (3) go for feature extraction of each subimage
Algorithm 2: (global zone)	For boundary image IM(, ); (1) calculate density in horizontal direction; (2) calculate density in vertical direction; (3) divide the density by number of bins in horizontal/vertical direction to calculate ; obtain the indices for zone boundary using; (4) obtain the subimages generated from IM using and; (5) go for feature extraction of each subimage
Algorithm 3: (local zone)	For boundary image IM(, ); (1) calculate density in horizontal direction; (2) calculate density in vertical direction; (3) divide the density and by number of bins in horizontal and vertical directions to calculate density for locating the boundary indices of zone and; (4) obtain the subimages generated from IM using only; (5) obtain for each density and divide this by the number of bins in horizontal direction to calculate density for locating vertical index of zone; (6) obtain subimage from using and; (7) go for feature extraction of each subimage