Dissertation > Excellent graduate degree dissertation topics show

A Study of Affective States Recognition Based on Ant Colony Optimization from Physiological Signals

Author: MaChangWei
Tutor: LiuGuangYuan
School: Southwestern University
Course: Signal and Information Processing
Keywords: Ant Colony Optimization Physiological signals Emotion Recognition Feature Selection
CLC: TP391.4
Type: Master's thesis
Year: 2010
Downloads: 87
Quote: 0
Read: Download Dissertation

Abstract


With the rapid development of computer technology, human-computer interaction is gradually being researchers. Affective computing is one of the topics currently emerging, the purpose is to give the computer recognition, understanding, emotional expression and the ability to adapt to people to enable them to perceive emotions, respond with timely and emotional interaction with the user. Affective Computing Research emotion recognition is one of the key issues, the main physiological signal emotion recognition step, the data acquisition and pre-processing and equipment were tested by a greater impact, feature extraction methods and classifier design is relatively mature, the degree of improvement is not large, and feature selection is worthy of our in-depth consideration. Feature selection is essentially a combinatorial optimization problem. ACO (Ant Colony Optimization, ACO) is a global optimization based on swarm intelligence algorithm, with coding is simple, fast calculation, population diversity, and easy to understand and implement other characteristics, is ideal for solving combinatorial optimization problems. Thesis ECG (Electrocardiography, ECG) and EMG (Electromyography, EMG) for the study, the local search and mutation strategy is introduced into ant colony optimization algorithm, combined with the introduction of incremental K, K-nearest neighbor classifier (K- Nearest Neighbor, KNN) for emotion recognition, access to the better results. Specifically to do the following tasks: (1) using wavelet transform R wave and S-wave amplitude and the RR interval and the absolute value of the information to locate the position of the QRS complex to extract statistical characteristics, accuracy rate of 99.5%. (2) for the physiological signals to identify affective characteristics combinatorial optimization problems, will introduce the idea of ??computational intelligence emotional physiological signals feature selection, the comparison of the ant system (Ant System, AS) and Ant Colony System (Ant Colony System, ACS ) two algorithms in the two sets of data collection under the recognition result: the local search strategy and mutation introduces ACS for feature selection, two emotions on joy and sadness classify a large extent improved the correct recognition rate and the speed of convergence; the ACS-specific pseudo-random proportion of maximum and minimum rules into ant system (MAX-MIN Ant System, MMAS), the same combination of local search and mutation strategy for feature selection, the recognition rate guarantee, while effectively feature combination. These classifiers are KNN. (3) For the laboratory test is being collected calm, happy, surprise, disgust, sadness, anger and fear seven different emotional state of the EMG signal, the feature extraction using IMMAS Fisher classifier combining feature selection from identified a variety of emotion an emotion. Simulation results from the preliminary experiment, we got some certain extent represent a combination of features emotion.

Related Dissertations

  1. Research on Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  2. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  3. Research on Feature Selection and Construction in Emotion Speech Recognition,TP18
  4. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  5. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  6. Improvement Ant Colony Algorithms and Its Application to Blind Equalization,TN911.5
  7. Research on Dynamic Routing and Wavelength Assignment Algorithms in Intelligent Optical Network,TN929.1
  8. Distributed target detection in wireless sensor network research,TP212.9
  9. FSVM -based data mining method and its application to intrusion detection research,TP393.08
  10. Key Technologies of WEB Media Resources Intelligent Discovery,TP393.092
  11. Ensemble of Multiple Descriptors for Automatic Image Annotation,TP391.41
  12. SMS User Interest Hierarchy Algorithm Based on Text Classification Algorithm,TP391.1
  13. Speech Emotion Recognition Based on Multifractal,TN912.34
  14. Italian group based text classification method,TP391.1
  15. The Research of Facial Expression Recognition Technology Based on GABOR Wavelet Transformation,TP391.41
  16. Lung Nodules CAD System Based on Chest CT Image,TP391.41
  17. Ant Colony Optimization and Its Application to Blind Equalization,TN911.5
  18. Feature Selection Based on Feature Extraction,TP391.41
  19. The Research on Method of Music Emotion Recognition,TP391.4
  20. A Novel Method of Nonlinear Rapid Feature Selection for High Dimension Features and Its Application on Bioinformatics,Q811.4
  21. Research and Development of Production and Scheduling System Base on IC Production Model,F273

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