Online Signature Verification Based on Multi-mode Matching
Yuhang Chen, Mahpirat Mahpirat, Kurban Ubul · 2020
In order to improve the accuracy of online signature verification, an online signature verification technology based on multi-mode matching is proposed. Combined with dynamic time warping (DTW) and string editing distance (SED), the average value, maximum value, minimum value and median value of the distance between the signatures to be tested and the signature of the template were calculated respectively. After normalization, it was combined into eight-dimensional feature vectors, and the experiment was carried out using SVM classifier. The experiment eventually achieved 1.25% FAR and FRR on SVC2004 Task2.