Recognition of facial expressions using self organizing feature maps and bidirectional associative memory | |
| Author | Surya, Djoni |
| Call Number | AIT Thesis no.CS-94-38 |
| Subject(s) | Neural networks (Computer science) Facial expression |
| Note | A thesis submitted in partial fulfilment of the requirements for the degree of Master of Science, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Abstract | Human face is a complex entity. It is difficult to describe its expression, because it contains vague and weak patterns at different parts of the face rather than a clear pattern at any one part of it. Bidirectional Associative Memory (BAM) is a way to represent vague concepts. Self organizing feature map is also used in order to obtain the weights that are further used as an input to BAM. We apply the new modified BAM learning method which can recognize many types of test patterns, including distorted ones. The BAM learning method is modified by augmenting a set of bipolar orthogonal patterns to the original patterns. |
| Year | 1994 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Sadananda, Ramakoti |
| Examination Committee(s) | Yulu, Qi ;Nagarur, Nagendra N. |
| Scholarship Donor(s) | Yayasan Gunadaima |
| Degree | Thesis (M.Sc.) - Asian Institute of Technology, 1994 |