Dissertation > Excellent graduate degree dissertation topics show
Image Feature Extraction for Face Recognition
Author: ZouJianFa
Tutor: WangGuoZuo
School: Chongqing University of Posts and Telecommunications
Course: Applied Computer Technology
Keywords: Face Recognition Feature selection Regression analysis Gabor feature Principal Component Analysis Linear discriminant analysis
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 135
Quote: 0
Read: Download Dissertation
Abstract
|
The face recognition technology is the use of computer analysis of face images to extract the effective identification information identifying or discriminating a pending state technology . It combines artificial intelligence , computer image processing , pattern recognition , psychology and other research fields , and is one of the hot spots in current research in artificial intelligence and pattern recognition . In the field of face recognition technology , what features to represent the face pattern , as well as how to effectively extract such features , often determines the good and bad effects of face recognition . In this paper, the image features Face Recognition , focus on how to extract and select features to face mode to carry out research work : First , principal component analysis (PCA) face recognition algorithm in feature selection exists proposed face recognition algorithm based on dynamic principal component molecular space . This method starting from the point of view of the image reconstruction using the theory of multiple linear regression analysis , the role of the principal component analysis , further screening has trained primaries feature sub space , thus removing the principal component of the characteristics of those interference identification results , select more conducive to the image to be identifying the characteristics of the main ingredients . The experimental results show that the algorithm than traditional based on the static main into the molecular space algorithm to identify the effect of significantly improved . Secondly , Gabor feature from a different direction and scale well represents local features of face images , but the traditional method using Gabor features but ignore the hominid face picture contains a wealth of global characteristics . Gabor features and the original image information is combined to form Gabor feature enhancements , and a combination of direct stepwise linear discriminant analysis algorithm (DF_LDA) , proposed a new face recognition method . Experimental results show that the algorithm has high recognition rate , steady performance . Finally , according to the needs of the research work , designed and implemented a face recognition system . The system implements several major face recognition method , and visualize the data extracted features demo . Meanwhile, in this system was added a strong interaction automatic face recognition module , the module to real-time video sequence from the camera acquisition , feature extraction , to achieve real-time face identification .
|
Related Dissertations
- Research on Algorithms of 2D Face Template Protection,TP391.41
- Research on Feature Selection and Construction in Emotion Speech Recognition,TP18
- Research of Video Face Recognition Based on Weighted Voting and Key-Frame Extraction,TP391.41
- Face Recognition Method Based on DE,TP391.41
- Based on local binary pattern and Weber local descriptor Face Recognition,TP391.41
- Expression of face recognition algorithm based on sparse research,TP391.41
- Under the conditions of variable lighting and occlusion of face recognition technology and its application,TP391.41
- Research on the Fault Diagnosis Technology of Reciprocating Compressors,TH45
- Study on Risk Factors of Gastroesophageal Reflux Disease,R571
- Study on Chinese Text Classification Combined with Ontology,TP391.1
- Text- based English blog space analysis of emotion,TP391.1
- Research on Face Recognition Based on Improved Maximum Margin Criterion,TP391.41
- Identified and articulated object detection applications,TP391.41
- Research on Speaker Recognition System under the Visual C++ 6.0,TN912.34
- Research on Subspace Analysis Based Face Recognition Algorithm,TP391.41
- Research on Network Traffic Classification Based on Decision Tree,TP393.06
- Attribute Reduction Based on Rough Set and Weighted SVM intrusion detection method,TP393.08
- Front-line police Burnout and Personality Traits Correlation,C912.6
- Yunnan urban-rural income gap and fiscal policies Empirical Study,F812.0;F224
- Peasant Workers Training Wishes and the Affecting Factors Analysis Based on the Investigation of Fuzhou,F224
- Research on Over-investment Behavior and Corporate Governance Mechanism,F224
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
|