Face recognition using a DCT-HMM approach
- 1 January 1998
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
A transform domain approach coupled with HiddenMarkov Model (HMM) for face recognition is presented.JPEG kind of strategy is employed to transform input sub-imagefor training HMMs. DCT transformed vectors of faceimages are used to train ergodic HMM and later for recognition.ORL face database of 40 subjects with 10 imagesper subject is used to evaluate the performance of the proposedmethod. 5 images per subject are used for trainingand the rest 5 for recognition. This method has an accuracyof 99.5%. The results, to the best of knowledge of the authors,give the best recognition percentage as compared toany other method reported so far on ORL face database.Keywords
This publication has 9 references indexed in Scilit:
- Finding faces in photographsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1998
- Face recognition/detection by probabilistic decision-based neural networkIEEE Transactions on Neural Networks, 1997
- Face recognition: a convolutional neural-network approachIEEE Transactions on Neural Networks, 1997
- Human and machine recognition of faces: a surveyProceedings of the IEEE, 1995
- Unsupervised texture segmentation using multichannel decomposition and hidden Markov modelsIEEE Transactions on Image Processing, 1995
- HMM-based architecture for face identificationImage and Vision Computing, 1994
- Object recognition using hidden Markov modelsMachine Intelligence and Pattern Recognition, 1994
- Eigenfaces for RecognitionJournal of Cognitive Neuroscience, 1991
- A tutorial on hidden Markov models and selected applications in speech recognitionProceedings of the IEEE, 1989