Dissertation > Excellent graduate degree dissertation topics show

The Research on Part-Based Object Recognition

Author: LiJianMin
Tutor: LiCuiHua
School: Xiamen University
Course: Applied Computer Technology
Keywords: Contour fragment chamfer distance object recognition mean shift machine learning
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 319
Quote: 3
Read: Download Dissertation

Abstract


Psychophysical studies show that human being can recognize objects using fragments of outline contour alone. Some other scholar also found that computer can recognize objects only based on the fragments of outline contour alone too. In this paper, we do systematic, deep and comprehensive research on the whole procedure of the part-based object Recognize and the technology related. And we get quite good result from the experiment.First we should get plenty of fragments of outline contour from picture samples, and build a codebook of Contour Fragments. In this paper, we propose a method of extracting fragments based on corner. Corner is a significant local feature of images, and contains the most shape information of the images. Because of the special Properties of the corner, the outline contour around them is the most typical fragments of the object. So we get the outline contour in the window which center is the corner as the fragment we need.Then we should select the most normal fragments from the fragments we got as the template for the training and recognition. We employ k-medoids clustering algorithm to clusters the fragments set into k clusters, and then select the center fragments to build the template. To this end, we employ the chamfer distance to measures the similarity of two contours at a certain relative location.Machine learning is an important and popular method in Artificial Intelligence. In this paper, we use the machine learning method to recognize the object. We try the AdaBoost and SVM method to train classifier, and make some comparisons. We compute the probabilities of the object position using the classifier above mentioned, and then employ mean shift method to find the max value. At last we complete the procedure of the recognition.

Related Dissertations

  1. Research on Intelligent Learning-Based Multi-Sensor Target Recognition and Tracking System,TP391.41
  2. Aerial Target Anti-interference Recognition and Tracking System,TN215
  3. Tracking Cells in High Density Image Sequences Based on Mean Shift Algorithm Combined with Topological Constraint,Q25
  4. The Research on English-Chinese Name Entity Translation,TP391.2
  5. Based on Data Distribution Characteristics of Text Classification,TP391.1
  6. Prediction of Binding Affinity of Human Transporter Associated with Antigen Processing,R392.1
  7. Research of Video-based Human-Computer Interaction Mode,TP391.41
  8. Research and Implementation of Multi-Scale Remote Sensing Image Segmentation,TP391.41
  9. Design and Implementation of Moving Object Tracking Algorithms Based on Mean Shift in the Visual Surveillance,TP391.41
  10. Research on Segmentation Algorithms of Regions of Interest for Pulmonary Lesions,TP391.41
  11. The Study of Fatigue Detection of Drivers Based on the Infrared Conditions,TP391.41
  12. Research and Implement of Chinese Word Segment Techniques Based on the Conditional Random Field,TP391.1
  13. Research on Key Technology in License Plate Recognition,TP391.41
  14. Research of Image Segmentation in Web Image Search Based on GPUs,TP391.41
  15. Research on Visual Tracking Algorithm Based on SIFT,TP391.41
  16. Research and Implement on Video Traking Technology Based on Kalman Filter,TP391.41
  17. Supervision topic model research and application,TP391.1
  18. Distortion effects on image quality evaluation and classification,TP391.41
  19. Social network algorithm and its application in recommendation,TP391.3
  20. SVM Based on SIFT and scene classification,TP391.41
  21. Heterogeneous Transfer Learning between Image and Text Data,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