Off line writer identification for Arabic language: Analysis and classification techniques using subwords features
Makki Maliki, Naseer Al‐Jawad, Sabah Jassim · 2017
Handwritten text in any language is believed to convey a great deal of information about writers' personality and identity. Handwritten documents are frequently used as evidences in forensic tasks. Handwriting skills is learnt and developed from the early schooling stages. Research interest in behavioral biometrics was the main driving force behind the growth in research into Writer Identification (WI) from handwritten text, but recent rise in terrorism associated with extreme religious ideologies spreading primarily, but not exclusively, from the middle-east has led to a surge of interest in WI from handwritten text in Arabic and similar languages. This paper is the main outcome of extensive research investigations conducted with the aim of developing an automatic identification of a person from handwritten Arabic text samples. For the dependent 8-dimensional WI scheme, we identify the best performing set of subwords (best 22 subwords out of 49 then followed by best 11 out of these 22 subwords). We established the validity of our hypothesis for different versions of subwords based Writer identification WI schemes by providing empirical evidence when testing on a number of existing text-dependent. The database consisted of 20 text samples from 95 writers. In these experiments we used two different similarity functions, one where we considered the Euclidian distance function while the other one combines the Dynamic Time Warping (DTW) for the projection attributes with Euclidian for the other features. The experimental results confirmed beyond any doubt the validity of our hypothesis and outperformed existing word-base WI, which only achieves high accuracy at rank 10 nearest neighbours. The 11 subwords based schemes with the DTW related similarity function achieved the higher accuracy in top rank around 98.12% accuracy rate of WI.