Character recognition using Radon Transformation and Principal Component Analysis in postal applications

This paper describes the method of handwritten characters recognition and the experiments carried out with it. The characters used in our experiments are numeric characters used in post code of mail pieces. The article contains basic image processing of the character and calculation of characteristics features, on basis of which it will be recognized. The main objective of this article is to use Radon transform and principal component analysis methods to obtain a set of features which are invariant under translation, rotation, and scaling. Sources of errors as well as possible improvement of classification results will be discussed.