Face recognition using Hidden Markov Models

[摘要] This thesis relates to the design, implementation and evaluation of statisticalface recognition techniques. In particular, the use of Hidden MarkovModels in various forms is investigated as a recognition tool and criticallyevaluated. Current face recognition techniques are very dependent on issueslike background noise, lighting and position of key features (ie. the eyes,lips etc.). Using an approach which specifically uses an embedded HiddenMarkov Model along with spectral domain feature extraction techniques,shows that these dependencies may be lessened while high recognition ratesare maintained.

[发布机构] Stellenbosch University