Human face profile identification | |
| Author | Nuanwan Soonthornphisaj |
| Call Number | AIT Thesis no. CS-97-13 |
| Subject(s) | Face perception Pattern recognition systems |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Advanced Technologies |
| Publisher | Asian Institute of Technology |
| Abstract | The research activities in the area of face recognition have grown significantly. Despite the high accuracy rate that has been achieved, there is still more research required in order to increase the speed used fo r recognition and to deal with face images take under more realistic conditions. Improved technique, based on geometric feature vector components, for human face profile identification had been proposed to get a better performance. Six fiducia l marks which are nose tip, chin, forehead, nose bottom, bridge and throat were proposed and used to construct the vector components. The vector components have three main groups which are angle, area, distance. The identification process based on the euclidean distance calculation . . Dealing with a variation of profile's angle was possible for the correct matching due to the effectiveness of a recheck procedure that was developed by this study. This procedure could increase the matching accuracy by reinvestigate each vector component of the candidate vector. The speed of execution time was also a significant improvement since less fiducial marks had been used to construct less vector components for the matching procedure. Finally, 98% for the correct matching could be achieved. |
| Year | 1997 |
| Type | Thesis |
| School | School of Advanced Technologies (SAT) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Yulu, Qi; |
| Examination Committee(s) | Sadananda, R. ;Murai, Shunji; |
| Degree | Thesis (M.Sc.) - Asian Institute of Technology, 1997 |