No abstract

Recognition of handwritten digits by computers is a common research topic in the pattern recognition area and has application in several domains. Many techniques can be applied in order to maximize the recognition performance, such as image preprocessing, feature extraction, feature selection and classification stages. This paper focuses on the assessment of three swarm intelligence optimization algorithms for feature selection optimization, called Binary Fish School Search (BFSS), Advanced Binary Ant Colony Optimization (ABACO) and Binary Particle Swarm Optimization (BPSO), for the recognition of handwritten digits. These meta-heuristics were applied to the well-known handwritten digit database MNIST, preprocessed with the CDF 9/7 Wavelet Transform. We used support vector machine (SVM) for the classification task. A considerable reduction in the number of features used for digit classification on the MNIST database with a small loss in the classification rates was observed.Keywords-feature selection; binary optimization; support vector machine; wavelet transform; handwritten digit recognition; binary fish school search; advanced binary ant colony optimization; binary particle swarm optimization.

Abstract. We present a novel preprocessing technique for handwritten numerals recognition, that relies on the extraction of multiscale features to characterize the classes. These features are obtained by means of different continuous wavelet transforms, which behave as scale-dependent bandpass filters, and give information on local orientation of the strokes. First a shape-preserving, smooth and smaller version of the digit is extracted. Second, a complementary feature vector is constructed, that captures certain properties of the digits, such as orientation, gradients and curvature at different scales. The accuracy with which the selected features describe the original digits is assessed with a neural network classifier of the multilayer perceptron (MLP) type. The proposed method gives satisfactory results, regarding the dimensionality reduction as well as the recognition rates on the testing sets of CENPARMI and MNIST databases; the recognition rate being 92.60 % for the CENPARMI database and 98.22 % for the MNIST database.