A Cascaded Scheme for Recognition of Handprinted Numerals U. Bhattacharya T. K. Das B. B. Chaudhuri CVPR Unit, Indian Statistical Institute, Kolkata, India ujjwal,das t,bbc @isical.ac.in Abstract This paper proposes a novel off-line handprinted Bangla (a major Indian script) numeral recognition scheme using a multistage classifier system comprising multilayer percep- tron (MLP) neural networks. In this scheme we consider multiresolution features based on wavelet transforms. We start from certain coarse resolution level of wavelet repre- sentation and if rejection occurs at this level of the classifier, the input pattern is passed to a larger MLP network corre- sponding to the next higher resolution level. For simplicity and efficiency we considered only three coarse-to-fine res- olution levels in the present work. The system was trained and tested on a database of 9000 samples of handprinted Bangla (a major Indian script) numerals. For improved generalization and to avoid overtraining, the whole avail- able data set had been divided into three subsets – train- ing set, validation set and test set. We achieved 94.96% and 93.025% correct recognition rates on training and test sets respectively. The proposed recognition scheme is ro- bust with respect to various writing styles and sizes as well as presence of considerable noise. Moreover, the present scheme is sufficiently fast for its real-life applications. 1. Introduction Off-line recognition of handwritten characters, in particular numerals has been a topic of intensive research during last few years. The application areas include postal code read- ing, automatic processing of bank cheques, office automa- tion and various other scientific and business applications. Automatic recognition of handwritten characters is diffi- cult because of variations in style, size, shape, orientation etc., presence of noise and factors related to the writing instrument, writing surface, scanning device etc. To sim- plify the recognition scheme, many existing methods put constraints on handwriting with respect to tilt, size, relative positions, stroke connections, distortions etc. In this paper we consider numeral characters written inside rectangular boxes of fixed size. Enough research papers are found in English [19], Chi- nese [20], Korean [10], Arabic [1], Kanji [22] and other languages. For example see the review in [17]. However, only preliminary work [16, 2, 3] has been done on a script like Bangla, the second-most popular language and script in the Indian subcontinent and the fifth-most popular language in the world. In the previous census, it was found that only 3% of the educated population of West Bengal, the major Bangla speaking state of India, knows a foreign language (mainly English). One of the important issues related to handwriting recog- nition is the determination of a feature set which is reason- ably invariant with respect to shape variations caused by various writing styles. To tackle the problem we have cho- sen a wavelet based multistage approach. Wavelet based ap- proach has been used for handwritten character recognition previously [21, 11, 9] but not in a cascaded manner used by us. In wavelet analysis, the frequency of the basis function as well as the scale can be changed and thus it is possible to exploit the fact that high frequency features of a function are localized while low frequency features are spread over time. Real life images are composed of large areas of similar in- formation but sharp changes at object edges. The biological eyes are more sensitive to object edges rather than minor de- tails inside. Thus, in many situations, wavelet based tech- niques are suitable for image processing tasks. Moreover, wavelet, as a problem-solving tool fits naturally with digital computer with its basis functions defined by just multipli- cation and addition operators – there are no derivatives or integrals. In this paper, a three stage system is proposed where fea- tures in the form of wavelet coefficient matrices at different resolution levels are considered at three different stages of the recognition system. MLP networks with different ar- chitecture is used as classifiers. In the initial stage, a nu- meral is subjected to recognition using the low-low part of the wavelet coefficient matrix as the feature set. If the input character is not classified at this level, it is passed to the next stage using wavelet coefficients of the next higher level of resolution. If the pattern is again rejected at this stage, at- tempt is made by the last stage where the next higher level of wavelet features are considered. In this scheme, depicted in Fig 1, feature vectors are ob- tained by convolving the Daubechies-4 wavelets [6] with a character image. Three MLP network architectures are trained using training sets at three coarse-to-fine resolution