Handwriting recognition has been
one of the most fascinating and challenging research areas in
field of image processing and pattern recognition in the recent
years. Offline recognition is performed on a scanned image of
handwriting and thus contains no temporal data. In general,
handwriting recognition is classified into two types as off-line
and on-line handwriting recognition methods. In the off-line
recognition, the writing is usually captured optically by a
scanner and the full text is available as an image therefore
it contains no temporal data. Some existing techniques are
fusion based segmentation method [7]. In this approach,
over segmentation of words from text based on pixel density
between upper and lower base line with multiple expert
base validation for character recognition and classification
has been developed. Slant and skew errors are neglected in
this approach. In K-nearest neighbor [9] technique, feature
extraction depends on Euclidean distance between testing
point and reference point. Which is used to calculate KNN
neighbor. This method could classify images containing single
characters. In N-gram [1], training is done on text corpus and
character can be recognized belonging to this corpus only.
We propose a system which could extract characters from
running text.