Dissertation > Excellent graduate degree dissertation topics show

Research of Facial Expression Recognition Algorithm

Author: ZhaoHong
Tutor: YeXiNing
School: East China University of Science and Technology
Course: Control Science and Engineering
Keywords: Maximum between threshold segmentation method Local binary patterns Support Vector Machine Adaboost ( algorithm)
CLC: TP391.41
Type: Master's thesis
Year: 2012
Downloads: 84
Quote: 0
Read: Download Dissertation

Abstract


The purpose of facial expression recognition is the expression of the face by the computer in case of non-contact automatic identification , further analysis of a person's feelings or emotional , is an important research topic in the field of intelligent man-machine interface . Because the issue has important research significance and enormous practical value and attracted wide attention from researchers at home and abroad . Expression recognition as a complete recognition system , mainly including : face detection , feature extraction and expression recognition three links . This article in the above three aspects made ??by the main research work have : (1) In the face detection section , this paper proposed the improved maximum between-class threshold segmentation method , select the threshold of the center pixel point and neighborhood pixel point of the ash the mean value as a reference threshold value , and then calculate the variance of the gray scale value of the image segmentation according to the threshold value , selected so that the variance value of the maximum reference threshold is the final threshold . And the method of segmentation results were compared with the results of the original method . This article (2) In the feature extraction link LBP operator texture feature extraction, and will be obtained through statistical LBP histogram as for identifying the characteristics of data. ( 3) the expression recognition stage , on the basis of in-depth study of SVM and Adaboost algorithm proposed a fusion algorithm . The fusion algorithm not only has the SVM algorithm to handle non-linear problems , Linear excellent characteristics , can not be separated on the problem and the small sample size problem Adaboost algorithm also has good performance in training weak classifiers . Thereby obtaining an ideal classification . (4 ) fusion algorithm for multi- classification , the characteristics of the DAG SVM method , based on the experimental results , considering the overall performance of organizations sorted structure clearly set for expression recognition classifier , and get better effect. (5) In this paper, JAFFE expression database as training samples to identify the effect of the above methods are verified by the experimental data can be seen , the proposed method has better recognition rate .

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. Research on Autamatic Music Structrue Analysis,TN912.3
  3. The Fatigue State Recognition of the Driver Based on Eye Detection,TP391.41
  4. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  5. Fault Diagnosis Method Based on Support Vector Machine,TP18
  6. Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
  7. Research on Predicting Intrinsic Disorder Protein Structure Based on Supervision Manifold Learning Algorithm,Q51
  8. Research for Infrared Image Target Identification and Tracking Technology,TP391.41
  9. Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
  10. Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
  11. The Research on Intrusion Detection System Based on Machine Learning,TP393.08
  12. Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
  13. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  14. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  15. Research on Subimage Selection and Mathching Method for Synthetic Aperture Radar(SAR) Target Recognition,TN957.52
  16. Fundus Image Segmentation Based on SVM and Template Matching,TP391.41
  17. Research and Realization of License Plate Character Recognition Algorithm Based on SVM,TP391.41
  18. Cloud Classification Based on Geostationary Meteorological Satellite Imagery,TP391.41
  19. The Studies on Some Improvements of the GA and Their Applications in SVM,TP18
  20. The Research and Implementation of the Fault Management System for Triple Play,TP315

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile