Litcius/Paper detail

A Survey of Offline Handwritten Signature Verification Based on Deep Learning

Yusnur Muhtar, Wenxiong Kang, Aliya Rexit, Mahpirat Mahpirat, Kurban Ubul

202213 citationsDOI

Abstract

Handwritten signatures are biometrics and a point of contention in the scientific community, the process used to verify whether a person’s signature is genuine. In the past ten years, the application of handwritten signature technology in the fields of administration, finance, handling legal disputes, and security has been greatly developed, and many researchers have focused on applying systems based on handwritten signature analysis and processing to new fields on the possibility. After several years of disorderly development in this field of research, it is time to assess the applicability of its current developments to formulate a structured path forward. In this paper, we provide a systematic review of the literature on offline handwritten signatures over the past 10 years, focusing on the most prominent and promising deep learning-based signature verification methods, and attempt to elicit possible future research directions on this topic.

Topics & Concepts

Signature (topology)Computer scienceBiometricsHandwriting recognitionField (mathematics)Process (computing)Artificial intelligenceSignature recognitionHandwritingPoint (geometry)Data sciencePattern recognition (psychology)Data miningFeature extractionPure mathematicsMathematicsGeometryOperating systemHandwritten Text Recognition TechniquesVehicle License Plate RecognitionNatural Language Processing Techniques