A new method for writer identification based on historical documents

Abstract Identifying the writer of a handwritten document has remained an interesting pattern classification problem for examiners, forensic experts, and paleographers. While mature identification systems have been developed handwriting in contemporary documents, remains challenging from viewpoint historical manuscripts. Design development expert that can identify questioned manuscript or retrieve samples belonging to given greatly help paleographers their practices. In this context, current study exploits textural information characterize documents. More specifically, we employ oBIF(oriented Basic Image Features) hinge features introduce novel moment-based matching method compare feature vectors extracted writing samples. Classification is based on minimization similarity criterion using proposed moment distance. A comprehensive series experiments International Conference Document Analysis Recognition 2017 dataset reported promising results validated ideas put forward study.