Research and Improvement on HMM-Based Face Recognition

Yan Fen Cheng; Hao Tang; Xian Qiao Chen

doi:10.4028/www.scientific.net/AMM.490-491.1338

Paper Titles

Analysis of Shock Wave Signals and Calculation of Trajectory Vector in Acoustic Positioning System
p.1317

The Data Link Layer Protocols for Underwater Acoustic Networks
p.1322

Processing Relational Top-N Queries with Text and Numeric Attributes
p.1326

AgCleaning: A Track Data Filling Algorithm Based on Movement Recency for RFID Track Data
p.1330

Research and Improvement on HMM-Based Face Recognition
p.1338

Managing Fragmented Database Replication for Mygrants Using Binary Vote Assignment on Cloud Quorum
p.1342

A Novel Weighted Dynamic Time Warping for Light Weight Speaker-Dependent Speech Recognition in Noisy and Bad Recording Conditions
p.1347

Seismic Data Denoising Simulation Research Based on Wavelet Transform
p.1356

Trajectory Pattern Mining: Methods and Applications
p.1361

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 490-491Research and Improvement on HMM-Based Face...

Research and Improvement on HMM-Based Face Recognition

Abstract:

This paper is concerned with face recognition using the hidden Markov model with 2D-discrete cosine transformation observations. The first part of the paper mainly discusses the influence of sampling parameter selection on model training and recognition efficiency and proposes method to increases the model efficiency through selecting optimal combinations of input parameters. In the second part of the paper, we choose the optimal parameters as input data standard for image matching and extend Viterbi algorithm by setting thresholds, the recognition time is reduced by 12.4% on average.