Please use this identifier to cite or link to this item:
|Title:||Human face recognition using line edge information||Authors:||Gao, Yong Sheng.||Keywords:||DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition||Issue Date:||2000||Abstract:||The automatic recognition of human faces presents a significant challenge to the pattern recognition research community. Typically, human faces are very similar in structure with minor differences from person to person. They are actually within one class of "human face". Furthermore, pose variations, lighting condition changes and facial expressions further complicate the face recognition task as one of the difficult problems in pattern analysis. This thesis has proved the proposed novel concept, "Faces can be recognized using Line Edge Map". A compact face feature, Line Edge Map (LEM), is generated for face coding and recognition. A complete investigation on the proposed concept is conducted according to two organization schemes. The first scheme covers all three levels (low, intermediate and high) pattern recognition methodologies that make use of information from spatial domain to structural and syntactic representations. The second scheme covers all aspects on human face recognition. The investigation covers face recognition under (a) controlled/ideal condition, (b) pose variation, (c) varying lighting condition, and (d) varying facial expression. The system performances are also compared with Eigenface method, one of the best face recognition techniques, and reported experiment results of other methods. It is a very encouraging finding that the proposed face recognition technique has performed consistently superior to (or equally well as) the Eigenface method in all the comparison experiments. The results demonstrate that LEM of human faces efficiently provide sufficient information for face recognition. LEM and the proposed generic Line Segment Hausdorff Distance measure provide a new way for face coding and recognition.||URI:||http://hdl.handle.net/10356/2632||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||SCSE Theses|
checked on Sep 25, 2020
checked on Sep 25, 2020
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.