An Offline Writer-Independent Signature Verification Method with Robustness Against Scalings and Rotations

Handwritten signatures are still one of the most used and accepted methods for user identification and authentication. They are used in a wide range of human daily tasks, including applications from banking to legal processes. The signature verification problem consists of verifying whether a given handwritten signature was generated by a particular person, by comparing it (directly or indirectly) to genuine signatures from that person. In this paper, we introduce a new offline writer-independent signature verification method based on a combination of handcrafted Moving Least-Squares features and features transferred from a convolutional neural network. In our experiments, our method outperforms state-of-the-art techniques on Western-style signatures (CEDAR dataset), while also obtaining good results on South Asian-style handwriting (Bangla and Hindi datasets). Furthermore, we demonstrate that the proposed method is the most robust in relation to differences in scale and rotation of the signature images. We also present a discussion on dataset bias and a small user study, showing that our technique outperforms the expected human accuracy on the signature-verification task.