Handwritten signature identification based on MobileNets model and support vector machine classifier

Biometrics is a field that uses behavioral and biological traits to identify/verify person. Characteristics include handwrittien signature, iris, gait, fingerprint. Signature-based biometric systems are common due their simple collection non-intrusive. Identify the humans using handwritten signatures has received an important attention in several modern crucial applications such as automatic bank check, law-enforcements, historical documents processing. Therefore, this paper accurate system proposed. The proposed preprocessing stage for input images. Besides, new deep learning model called MobileNets, which used classification process. Support vector machine (SVM) classifier with MobileNets inorder get better identifaction results. Experimental results conducted on standard CEDAR, ICDER, sigcomp signature datasets report 99.8%, 98.2%, 99.5%, identification accuracy, respectively.