A Perspective Analysis of Handwritten Signature Technology
ACM Computing Surveys
Handwritten signatures are biometric traits at the center of debate in the scientific community. Over the last 40 years, the interest in signature studies has grown steadily, having as its main reference the application of automatic signature verification, as previously published reviews in 1989, 2000, and 2008 bear witness. Ever since, and over the last 10 years, the application of handwritten signature technology has strongly evolved and much research has focused on the possibility of
... systems based on handwritten signature analysis and processing to a multitude of new fields. After several years of haphazard growth of this research area, it is time to assess its current developments for their applicability in order to draw a structured way forward. This perspective reports a systematic review of the last 10 years of the literature on handwritten signatures with respect to the new scenario, focusing on the most promising domains of research and trying to elicit possible future research directions in this subject. M. Diaz et al. that they contain the temporal and dynamic order in which the signer executed the signature. To register this kind of signature, a device such as a digitizing tablet is required. Although all the ASV systems operate according to the general scheme illustrated in Figure 1 , the strategy for extracting the features and producing the final classification is generally different between online and offline systems. Nevertheless, to test an ASV, a meaningful statistical signature database is required in both cases. The article is organized as follows: Section 2 discusses aspects related to handwritten signature databases, which have become a crucial aspect of the research. Best practices in automatic signature verification are discussed in Section 3. In Section 4, a summary of the main competition results is reported by highlighting the state of the art in the field. The forensic aspects related to signature analysis are discussed in Section 5. Some of the most promising new research directions are addressed in Section 6. Section 7 concludes on a positive and optimistic note regarding the future of this technology. DATABASES Handwritten signatures are the product of a complex human process. There are, therefore, innumerable variables and parameters that cannot be represented in a single signature database of a practical size. ASVs typically model the peculiarities in the signatures found in a particular signature database, and this limits their applicability. This effect probably leads to biased results since not all human behavior is taken into account in the datasets. As such, the challenge is not only to improve the state-of-the-art results with current databases but also to develop incremental signature databases, which consider more and more examples of human behavior: temporal evolution, aging, posture, emotions, multiscripts, collection devices, user skills in using new devices, occlusions (e.g., lines, printed or handwritten text or stamps), and so on. As more aspects and conditions of human behavior are considered, the ASV algorithm is tested more realistically. For a reliable test of current and forthcoming algorithms, large projects to collect huge handwritten signature databases or open government data initiatives are needed.