A New Database for Writer Demographics Attributes Detection Based on Off-Line Persian and English Handwriting
Alieh Masomi, Hamid Reza Ghafari, Kazem Nouri, Younes Akbari, Walid Bouamra, Chawki Djeddi · 2016
This paper describes a database of multi-script (Persian and English) for typical and new aspects and challenges of offline handwriting automatic analysis field. This database can be used for typical aspects such as different levels of segmentation and recognition and writer identification in text-dependent and text-independent modes. Also, new aspects can be used such as writer identification and gender, age and handedness detection based on script-dependent and script-independent. Two pages of forms in three regions were designed and collected from 200 native Persian writers of different age, handednesses, genders and education level. To the best of our knowledge, so far no attempt has been conducted on providing Persian database for writer identification and gender, age and handedness detection based on script-dependent and script-independent in multi-script environment.