In this research, we offered a new and simple way of
Handwriting Characters Recognition. This way extracts positions of
the black points from binary images (black, white) according to
certain coordinates which are used in the stages of training an
d
testing. The extracted positions are stored in a database according
to appropriate structure for predictive data mining.
We used training data to build a predictive model which helps
in Recognition testing data depending on the data stored in the
database. We have conducted a number of tests on different
samples of handwriting character images. We got accurate results,
within the required conditions.