Dissertation > Excellent graduate degree dissertation topics show

The Research of Image Matching Method Based on Feature Descriptor

Author: TangLiNa
Tutor: KongJun
School: Northeast Normal University
Course: Applied Computer Technology
Keywords: Image Matching Feature Extraction Characterization SIFT Local Binary Pattern
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 98
Quote: 0
Read: Download Dissertation

Abstract


Currently, the method of acquiring images and more rich, get the number of images are growing in recent years, digital image processing research work has been a lot of attention of scholars, some image processing basic research are becoming important together. In many applications, the digital image as a new generation of digital information which is applied, it is more images need to be processed. Image matching as image processing a basic research direction has been quite widespread concern. Image matching problem both in target tracking, artificial intelligence, computer vision, face recognition or in other application areas, is one of the key issues that need to be resolved, however, to solve this critical problem is an effective method is to use some contain images Important structural information to create the image feature point transformation between between. However, this method has some difficulties, such as how to accurately extract sufficiently stable feature points, and build on the light, scale, obscure, rotation, noise and perspective transformation, etc. have a good robustness characterization child. It can be seen in the feature extraction and description of computer vision and image processing is the basic link, and image processing efficiency and accuracy of the results are processed by the feature detection operator detection performance and characterization of operator performance Biaozhen direct decision. How to choose the right image feature descriptor to make these features not only has good hands performance, but these changes can be maintained at constant has become a critical issue. In this paper, the existing image invariant feature detection and feature-based image matching algorithm described were analyzed and summarized, a fast rotation invariant image matching algorithm. Using traditional SIFT feature point detection operator, by setting reasonable parameters to ensure the stability and the number of feature points, and then the feature points around the neighborhood to build a local invariant feature rotation invariant feature descriptor. In this paper, the local invariant feature is the nature of a rotation invariant local binary patterns centrosymmetric (Rotation Invariant CS-LBP), it is through the traditional Local Binary Pattern (LBP) to modify get. After getting characterization vector, the use of appropriate methods to determine the similarity measure between pairs of matching points between two images in order to achieve a match. The new method and the conventional methods SIFT experiments on different data, by comparing the experimental results, we can see that the new method has better performance, especially in the presence of the case of a large angle of rotation.

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. The convergence of mathematics teaching in primary and secondary schools under the new curriculum with the problem representation,G633.6
  3. ISAR Imaging Simulation of Space Targets and Target Recognition Based on ISAR Images,TN957.52
  4. Research on Feature Extraction and Classification of Pulse Waveform for Cholecystitis and Nephrotic Syndrome Diagnosis,TP391.41
  5. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  6. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  7. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  8. Research on the Image Real-Time Acquisition, Storage and Image Processing System,TP391.41
  9. Multi-currency Notes Technology Research and Implementation,TP391.41
  10. The Research on Paper Currency Classification Method Based on Harr-Like Feature and Minimal Ball Including Samples,TP391.41
  11. Research on Visual Detection and Tracking of Mobile Robots,TP242.62
  12. Studies on Synthesis, Characterization and Properties of Fe-V Co-doped TiO2 Catalyst,O614.411
  13. Characterization of Corn Lectin and Application to Screen of Growth-promoting Bacteria from Corn Rhizosphere,S513
  14. Health-based and Ecological Risk Assessment of Contaminated Sites,X820.4
  15. Tobacco Diseases Auto-Recognition Research Based on Image Processing Technology,S435.72
  16. Preliminary Research on Characterization of Thermophilic Protease Pasey and Helicase Htc16,Q814
  17. Cloning and Expression of Xylanase Gene from Thermobifida Halotolerans YIM 90462~T,Q78
  18. Research on Nondestructive Detection Technology for External Qualities of Papayas Based-on Vision,S667.9
  19. Isolation and Characterization of Carbendazim-Degrading Strains, Cloning and Expression of the mheI Gene,X172
  20. Characterization of an Atrazine-Degrading Strain, Cloning of Key Degrading Related Genes and Construction of a Gene Cluster,X172
  21. Research on Identification System of Cashmere and Wool Fiber,TS101.921

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