Dissertation > Excellent graduate degree dissertation topics show

Study on Some Key Issues of Ear Recognition

Author: TianYing
Tutor: YuanZuoZuo
School: Shenyang University of Technology
Course: Measuring Technology and Instruments
Keywords: Biometric Ear Recognition Outer ear split Edge Detection Feature Extraction Geometric characteristics Scale Invariant Feature Transform Gabor wavelet Feature Fusion Force Field Transformation
CLC: TP391.41
Type: PhD thesis
Year: 2008
Downloads: 441
Quote: 2
Read: Download Dissertation

Abstract


Biometric technology is a unique human physiological or behavioral characteristics to personally identifiable, it provides a high reliability, high stability authentication ways. Ear Recognition is a new branch in the field of biometric identification, is also one of the challenging issues in the field of computer vision and pattern recognition. As the human ear is special physiological location and structural features of the ear recognition has potential applications in the field of public safety, information security, a growing cause for concern. Ear Recognition Research also in the experimental testing phase to develop a truly robust, practical ear recognition system, also need to solve many problems. In this paper, the problem of ear recognition more in-depth research, the main work of the following aspects: 1) an overview of the development of the human ear recognition, status and prospects. Feasibility of biometric identification technology for the human ear, the contents of the study, advantages and defects, a comprehensive overview of the human ear recognition method, a summary of the main problems and ear recognition. 2) to study the problem of image pre-processing of the human ear. A human ear image segmentation method based on template matching and genetic algorithm can be split from the the 2D grayscale man side face image on the human ear is obtained contains only the smallest rectangular region of the human ear. 3) to study the problems of the human ear image edge detection. First, according to the position and shape of the characteristics of the human ear to ear outer contour detection method based on boundary tracking people. Tracking can accurately extract the entire outer ear contour curves in accordance with the different movements of the auricle boundary contour of the external ear. Then, according to the characteristics of the human ear edge to edge detection method based on surface principal curvatures, on this basis, in order to detect the weak edges proposed multiscale edge detection method based on Hessian matrix can extract the curve of the edge of the human ear. The experimental results show that these methods can effectively extract the edges of the image of the human ear. 4) studied the problem of the alignment of the contour curve. According to the characteristics of the outer contour of the human ear curve curve registration method based on improved Hausdorff distance of the outer ear contour. The method rotation change the position changes and the plane of the image of the human ear is very robust, in order to further improve the recognition rate, and then use the generalized characteristic point of the inner ear for accurate identification. 5) invariant characteristics in recognition of the human ear. Scale Invariant Feature Transform (SIFT) can extract the stable characteristics of the key points and robust feature descriptor vector image of the human ear. Combined with Gabor wavelet technology, the use of spatial locality and orientation selectivity of Gabor wavelet excellent characteristics, the structure of low-dimensional point feature descriptor vector Gabor-SIFT, the premise does not change the recognition accuracy can effectively improve the matching efficiency. In addition, the operator for local feature matching defects fusion identification method based on SIFT features and geometric features can eliminate the ambiguity created an image has multiple local area similar illumination changes and some rotation angle variation of the human ear image recognition and achieved good results. 6) Force Field Transformation Theory and its application in ear recognition. The human ear feature extraction method based on force field transformation theory. According to the characteristics of the human ear, the force field were applied to the the ear image after the ear images and via energy conversion the auricle-contour shape of the feature point and the internal structure of the feature point can be extracted, the extracted feature with the stability, unique and strong The distinction between the ability of image rigid body transformation has invariance, illumination changes can effectively eliminate ear recognition to solve the problem of the low recognition rate local deformation caused by the attitude change and head deep rotation.

Related Dissertations

  1. Research on Multi-Modality Image Matching Technology Based on Gross Contour,TP391.41
  2. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  3. The Compression and Fusion Technique Research of Underwater Target Feature,TN911.7
  4. Research on Methods of Fingerprint Image Segmentation,TP391.41
  5. Face recognition algorithm based on feature fusion research,TP391.41
  6. Based on Information Fusion Fault Diagnosis of Analog Circuits,TN710
  7. Target Recognition Technology Based on Feature Fusion,TP391.41
  8. Palmprint Recognition Based on Feature Fusion,TP391.41
  9. The Research for Visual Surveillance System Based on Fusion of Multiple Model Data,TP391.41
  10. Research on Face Recognition Technique Based on Canonical Correlation Analysis,TP391.41
  11. A Study on Automatic Annotation of Thangka Image Based on Graph Learning,TP391.41
  12. Particle filter based video object tracking method,TP391.41
  13. Study of Method for Gait Recognition Based on the Fusion of Angles and Body Silhouette Feature,TP391.41
  14. The face recognition method based on canonical correlation analysis,TP391.41
  15. Research on Video Object Tracking Algorithm Based on Particle Filter,TP391.41
  16. Integration into high-level semantic feature of medical image retrieval technology,TP391.3
  17. Features and relevance feedback based on the underlying medical image retrieval,TP391.41
  18. Researches on 3d Model Retrieval Based on XML Technology and Semantic Matrix,TP391.41
  19. Research on Color Face Recognition Based on Fusion of Components Feature Extracted by Spectroface and FLD,TP391.41
  20. Based on face and fingerprint multi- modal identification system research,TP391.41

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