Online Kanji Characters Based Writer Identification Using Sequential Forward Floating Selection and Support Vector Machine

Writer identification has become a hot research topic in the fields of pattern recognition, forensic document analysis, criminal justice system, etc. The goal this is to propose an efficient approach for writer based on online handwritten Kanji characters. We collected 47,520 samples from 33 people who wrote 72 handwritten-based characters 20 times. extracted features handwriting data and proposed support vector machine (SVM)-based classifier identification. also conducted experiments see how accuracy changes with feature selection parameter tuning. Both text-dependent text-independent were studied work. In case identification, we obtained each character separately. then by considering some top discriminative case. Finally, another experiment was performed taking two, three, four instead using only one character. experimental results illustrated that SVM provided highest 99.0% 99.6% hope study will be helpful