Handwriting-based writer identification with complex wavelet transform
基于复小波的脱线手写体笔迹鉴别

A challenging problem of off-line text-independent writer identification is that plenty of dynamic writing information with the handwriting images can not be extracted as writing features, this results in high error rate in off-line writer identification. In order to enhance the performance of off-line writer identification, a complex wavelet-based Generalized Gaussian Distribution (GGD) method was proposed. Compared with the traditional wavelet-based GGD method, the novel method is more efficient on texture extraction due to its time-invariant features and good directional analysis. Experimental results show that the proposed method achieves a better performance of writer identification.