International Journal of Soft Computing and Engineering (IJSCE) Volume-1, Issue-1, March 2011 1 Abstract—Handwritten Numeral recognition plays a vital role in postal automation services especially in countries like India where multiple languages and scripts are used .Because of intermixing of these languages; it is very difficult to understand the script in which the pin code is written. Objective of this paper is to resolve this problem through Multilayer feed-forward back-propagation algorithm using two hidden layer. This work has been tested on five different popular Indian scripts namely Devnagri, English, Urdu, Tamil and Telugu. Network was trained to learn its behavior by adjusting the connection strengths on every iteration. The resultant of each presented training pattern was calculated to identify the minima on the error surface for each training pattern. Experiments were performed on samples by using two hidden layers and the results revealed that as the number of hidden layers is increased, more accuracy is achieved in large number of epochs Index Terms— Numeral Recognition, Artificial Neural Network, Supervised learning, Back Propagation Algorithm. I. INTRODUCTION Numeral Recognition refers to the process of translating images of hand-written, typewritten, or printed digits into a format understood by user for the purpose of editing, indexing/searching, and a reduction in storage size. The Numeral Recognition process is complicated by noisy inputs, image distortion, and differences between typefaces, sizes, and fonts. Numeral Recognition becomes more complex when multiple scripts were used during pin code or phone number writing because it is not possible for a single person to have an idea of every scripting of numeric digits. For solving this problem, artificial neural networks are commonly used. The software developed in this project converts numeric image written in five different scripts (Devnagri, English, Urdu, Tamil and Telugu) in to English so that it can be understood easily. Previously, different methods were proposed [1],[2],[3] for identifying multiscript numerals but still accuracy percentage is not satisfactory. The software used in this project uses two hidden layer so as to get more precise results. Experiments were performed on different multi scripting samples and 96.53% accuracy is achieved. Manuscript received Feb 22, 2011. Stuti Asthana is M.Tech Scholar from RKDF Institute of Science and Technology, Bhopal, India , e-mail: stutiasthana@gmail.com. Farha Haneef is M.Tech Coordinator in RKDF Institute of Science and Technology, Bhopal, India.e-mail: haneef_farha@yahoo.com. Rakesh K Bhujade is with Technocrats Institute of Technology, Bhopal , India., e-mail: rakesh.bhujade@ gmail.com. At present postal sorting machines are available in several countries like USA, UK, Canada, Japan, France, Germany etc. There are only a few works on Indian postal system [4], [5] and at present no postal automation machine is available for India Indian pin-code (postal code) is a six-digit number and system development towards Indian postal automation is more difficult and challenging than that of other country because of its multi-lingual and multi-script behaviour. In India there are 22 official languages and 11 scripts are used to write these languages. Fig. 1 shows a system that can substitute for the manual sorting letters. . Fig 1: Automatic letter sorting system on the basis of pin code Rest of the paper is organized as follows. Overview of Devnagri, English, Urdu, Tamil and Telugu scripts in section 2, phases in recognition of pin code numerals in section 3,and experimental results and comparative study in section 4 . II. OVERVIEW OF TAMIL, TELUGU, URDU, DEVNAGRI AND ENGLISH SCRIPTS REVIEW STAGE Tamil is a Dravidian language, and one of the oldest in the world. It is the official language of the Indian state of Tamil Nadu; it also has an official status in Sri Lanka, Malaysia and Singapore. The Tamil script has 10 numerals, 12 vowels, 18 consonants and five grantha letters. Telugu is a Dravidian language and has the third most popular script in India. It is the official language of the Indian state of Andhra Pradesh. There are 10 numerals, 18 vowels, 36 consonants, and three dual symbols. Urdu digits mostly followed in Pakistan Generally, in Urdu both Urdu numerals Handwritten Multiscript Numeral Recognition using Artificial Neural Networks Stuti Asthana, Farha Haneef, Rakesh K Bhujade Artificial Neural Network for recognizing the postal code Automatic Bin Sorter Robot Vision Conveyor belt