A neural network based efficient technique for Handwritten digit recognition process

data mining

This paper describe the hand scripted number and character recognition process which is developed using feed forward back propagation neural network. There is various platform like record rooms, where it demand for handscripted alpha numerals recognition which make the work easy and fast. Unlike the other neural network, feed forward back propagation net handle errors effectively by propagate them towards next step. At next step processing is execute with considering that previous exception, hence at next subsequent step error is minimizes and output is become more accurate. This application allows the user to write alpha numerals in there handscript style, shape and outfilm, and system identify the accurate scripted character or number. This project application may used in record room, banks, administrative offices and various more.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.