The task of recognizing handwritten numbers, using a classifier, is of great importance. This article applies the Radial Base Function technique for numerical handwriting recognition of Devanagari Script. Much work has been done in Devanagari numeral recognition using different techniques to increase recognition accuracy. Since the database is not created globally, the database was first created by implementing the preprocessing in the training data set. Then, through the use of Principal Component Analysis, we have extracted the characteristics of each image, some researchers have also used the extraction of density characteristics. Since different people have different writing styles, so here we are trying to form a system where numeral recognition becomes easy. Then, in the hidden layer, the centers are determined and the weights between the hidden layer and the output layer of each neuron are determined to calculate the output, where output is the adder value of each neuron.
Handwritten character recognition has existed since 1980. To date many inquiries have been made. Automatic reading of numeric fields has been attempted in various application areas, such as online hand recognition on computer tablets, recognition of postal codes in the mail for postal address sorting, bank check processing, Numerals in manually filled forms (for example,) and so on. In solving this domain of handwriting recognition many challenges are faced. Since handwritten digits do not always have the same size, thickness, or orientation and position in relation to the margins, many handwritten versions are even difficult to recognize.
The related video is here: