Human identification by handwriting of identity text
| Title | Human identification by handwriting of identity text |
|---|---|
| Summary | Identify the author of a hand-writing when repeated identity relevant text is available |
| Keywords | |
| TimeFrame | |
| References | |
| Prerequisites | |
| Author | |
| Supervisor | Josef Bigun |
| Level | Master |
| Status | Open |
Humans have been using hand-writting on paper or similar media to convey all sorts of information since time immemorial. A text type that we produce frequently is identitification text which contains identitity relevant information, e.g. name, date of birth, city of birth, city of residence, etc. Because we produce such text often, we write them with greater confidence and speed than other text, suggesting that there is more significant identity unique information in them, even if the text is analyzed offline. Additionally, some portions of such hand-written information are deliberately identity-unique, e.g. signature.
The need is among others caused by legal requirements. For example there can be thousands of participants who will simultaneously take a multiple choice exam, quiz, and each participant must enter her/his identity information and/or signature by hand on the response-forms.
The project will work with the quiz scenario in mind, i.e. the participants are students who take quizes at different times and for different courses. They will enter the same/similar identity information by hand-writing on answer forms. The order of incoming response-forms to the recognition system is random, because at each quiz the students sit at different places, and not all take the same quizes, etc. However, the number of persons taking the quiz is limited, e.g. the students of a school taking different courses.
The purpose of the project is to extract identity relevant information from repeated instances of text entered by hand-writting automatically using image processing techniques. The first goal is to tell which quizes are hand-written by the same student. The secondary goal is to extract the explicit identity of each student.
