Dissertation > Excellent graduate degree dissertation topics show

Research of Image Retrieval Algorithm Based on Fuzzy Matrix Learning

Author: GuoShiHui
Tutor: YangMing
School: Southwestern University
Course: Applied Computer Technology
Keywords: CBIR relevance feedback fuzzy semantic relevance matrix long-term learning
CLC: TP391.3
Type: Master's thesis
Year: 2012
Downloads: 53
Quote: 0
Read: Download Dissertation

Abstract


With the Development of Multimedia and Network Technology, There Are A Large Number of Images in Our Life. Although the Internet Provides Us with An Inexhaustible Multimedia Information Database, the Problem Is How to Find the Images of User’s Satisfaction Fast and Accurately, It Is A Difficulty Problem of Image Retrieval Currently. So There Is An Urgent Need for Effective Image Retrieval in Many Fields.Content-based Image Retrieval (Cbir) Has Been Proposed and Widely Used the Past Several Years. But the "Semantic Gap" Between the Low-level Features of An Image and High-level Semantic Feature Has Became the Major Obstacle to Cbir Technology. Relevance Feedback Mechanism Narrow the "Semantic Gap" to Some Extent.in This Paper, We First Comprehensively Discuss the Key Technology of Cbir and Relevance Feedback, Then, Made A Deeply Research to Solve the "Semantic Gap", and Provides A High-efficient and Practical Semantic-supported Image Retrieval System. the Main Work of This Paper Include:1. Combine of Color, Texture, Shape and Color-location Feature, It Provide Multi-choice to the User in Order to Meet the User’s Requirement.2. A Relevance Feedback Algorithm Based on Fuzzy Semantic Relevance Matrix Is Proposed. It Update the Value of Fsrm According to User’s Feedback, Which Capture the User’s Retrieval Intention. Through Learning the Value of Fsrm Constantly, and Then Update It to Bridge the "Semantic Gap", Finally, to Improve the Accurate of Retrieval. Experimental Results Prove the Effectiveness of the Algorithm.3. in Order to Improve the Time Performance of Retrieval and Feedback. the Corel Image Database Is Classified Using the Svm. the Method for Svm-based Image Classification Using Color Features. the Experimental Results Show That the Superiority of the Feedback Time on the Basis of the Classification.4. the Learning and Long Learning Algorithm Based on Fsrm Is Proposed. the Learning algorithm can effectively extract the latent semantic information, and update the value of FSRM according to the formula, and the semantic information of FSRM can spread faster to improve the accuracy of feedback. The experimental results prove the efficient of the algorithm. The long learning algorithm is based on the limited feedback learning, then analyze the semantic information of FSRM, and the wrong image in the image class is deleted to further improve the accuracy of feedback.5. Design and implement the retrieval system in this paper with Matlab7.8, it has the function of human-machine interactive. A series of experiments prove the effectiveness of this paper’s algorithm.

Related Dissertations

  1. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  2. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  3. The Application of Reinforcement Learning and Relevance Feedback in Orthodontics Image Retrieval,TP391.41
  4. Combining Semi-supervised and Active Learning for Image Retrieval,TP391.41
  5. Design and Implementation of Semantic Annotation System for 3D Models,TP391.41
  6. Research on 3D Model Retrieval System Based on Multi-feature and Relevance Feedback,TP391.3
  7. The Research on Algorithms of Relevance Feedback and Classification in Image Retrieval,TP391.41
  8. Research on Query Expansion Technique of Retrieval System in Biomedical Field,TP391.3
  9. Research on Algorithm Analysis and Applications of Manifold Learning,TP18
  10. Content -based image retrieval relevance feedback technology research,TP391.41
  11. Research on Key Technology of Internet Search Keywords Classification,TP391.1
  12. Content-based Image Retrieval Methodology and Implementation,TP391.41
  13. Research on Relevance Feedback in Vector Graphics Retrieval,TP391.41
  14. Retrieval Technology Research of Calcification Lesions in Mammograms,TP391.41
  15. Based image retrieval technology,TP391.41
  16. Research on Image Retrieval Technology Based on Texture Spectrum Feature,TP391.41
  17. Research of Image Retrieval Based on Relevance Feedback,TP391.41
  18. Research on Automatic Abstraction Based on Search Engine Result,TP391.1
  19. Research of Image Annotation and Retrieval Based on SVM,TP391.41
  20. Research and Implementation of SVM -based image retrieval system,TP391.41
  21. Research and application of content-based image retrieval,TP391.41

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