| Hi-index | 0.00 |
The recognition of two touching numerals has been tackled by many researchers with the purpose of recognizing the terminal fields in many document forms. The conventional methods are based on the process with two sequential stages, i.e., segmentation of touching numerals and recognition of the individual numerals. Due to an unlimited number of different overlapping and touching types, the segmentation-based approach has always revealed a limitation in success rates.In this paper, we propose a new segmentation-free method using neural network. In this approach, two touching numerals are regarded as a single pattern coming from a pattern source with 100 classes. To obtain a training set for the neural network classifier, we synthesize the patterns by moving two isolated numerals in the NIST database horizontally until they touch. For the test set, we manually extract two touching numerals from the numeral string dataset of NIST database. By using a modular neural network classifier, a promising result has been obtained.