Content Independent Writer Identification Using Occurrences of Writing Styles for Bangla Handwritings

In most forensic cases handwritten documents are useful for person identification. It is not easy to identify a writer or narrow down a writer from a group of writings. Each person has his or her own individual handwriting styles. This paper presents an approach to extract unique features based on writers' writing styles from the handwritten Bangla scripts. The computed features are used to identify the writer from a group of writings. An identification strategy using those features and rejection criteria are shown. The proposed approach is applied on a collected dataset and it gives us an encouraging results.