Recognition of handwritten similar Chinese characters based on support vector machines and partial space method

摘要

This paper presents a recognition approach for handwritten similar Chinese characters based on wavelet transform and elastic meshing technique, partial space method and support vector machines (SVM) theory. The technique of combining wavelet transform with elastic meshing is employed for a given handwritten character to extract its feature. In addition, partial space method is adopted to select some components of the feature vector showing the main difference among similar Chinese characters. Finally, SVM classifier based on the Gaussian kernel function is exploited to obtain the recognition result. The experiment results confirm the effectiveness and practicality of the approach.