Dissertation > Excellent graduate degree dissertation topics show

Rotational motion blurred image the analog and recovery methods research in non-ideal conditions

Author: XueKun
Tutor: ZhuJin
School: Nanjing University of Technology and Engineering
Course: Applied Computer Technology
Keywords: Rotational motion blur Blurred image simulation Image restoration Blind Restoration
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 108
Quote: 1
Read: Download Dissertation

Abstract


When shooting target and camera movement at the moment of exposure memory will cause the captured image is blurred, fuzzy known as motion blur. In the actual process, the motion between the camera and the object is relatively complex, and working on the motion platform, the camera shooting the target often accompanied by a non-coaxial rotating and oscillating on the motion platform. In this paper, for the rotational motion under non-ideal conditions blurred image simulation and restoration algorithms, specific work mainly in the following aspects: 1 In this paper, the method of analytic geometry and computer graphics research camera assumed non-ideal conditions (such as camera on a moving platform to an angle relative to the target for rotary motion, at the same time along with the swing motion platform) under the mathematical process to establish the model of image degradation. On the basis of this study, the rotational angular velocity, the swing angular velocity, the exposure time, the field angle of view, image size as a parameter, and rotating motion blur generated in the combination of different parameters the analog image. 2 In this paper, the rotary motion of the unknown parameters of fuzzy proposed a fuzzy parameter estimation methods: the rotational motion blurred image is divided into m × n small diagram, each fuzzy path fuzzy path can be approximated as a straight line. The paper uses the straight-line motion blur direction and fuzzy scale estimation method to identify the fuzzy scale of each of the small picture and direction, eventually forming a rotational motion blur field an estimate of the (m × n size). Then this article using the least squares fit to the the rotating motion blur Center and fuzzy angle. In addition, this estimation method for the rotating motion blur field from multiple angles blurred image estimated experimental, and the experimental results are compared and analyzed. Experimental results show that: Fuzzy a priori knowledge of the angle when the rotation center of a priori knowledge between the true values ??of plus or minus 20 pixels, the center of rotation of the least squares estimate and the true value of the error of plus or minus 10 pixels; when the error is within 5 °, fuzzy angle of least squares estimate is consistent with the true value of the error is small. In this paper, based on polar coordinates transform rotary motion blur restoration method, recovery experiments for the blurring of the image of the center of rotation in the center of the image. The experimental results show that: the the algorithm 5 ° and 10 ° rotation fuzzy fuzzy picture could better restore rotation fuzzy angle increases, restore the image effect is worse.

Related Dissertations

  1. Research on Restoration of Images Collected by Reconnaissance System in Near-Space,TP391.41
  2. A Video Inpainting Method Based on K-Dimension Tree Structure and Approximate Nearest Matching in Time-spatial Space,TP391.41
  3. Applications of Regularization Method in Image Restoration,TP391.41
  4. Image-Based Virtual Space Construction and Its Application,TP391.41
  5. Research on a Single Image Motion Deblurring,TP391.41
  6. Several Methods of Image Restoration,TP391.41
  7. PDE patch technology - based image coding research,TN919.81
  8. Image Restoration Based on Intelligent Algorithm,TP391.41
  9. Image Restoration Research on Frequency Domain and Wavelet Domain,TP391.41
  10. Millimeter wave passive imaging system modeling and blind recovery processing algorithms,TP391.41
  11. Image Inpainting Using Bandelets and PDE,TP391.41
  12. Learning-Based Restoratioin for Compressed Images,TP391.41
  13. Super- resolution algorithm for passive millimeter wave imaging studies and DSP Implementation,TP391.41
  14. Research of Motion Estimation Algorithm Based on Digital Images,TP391.41
  15. Research on Millimeter-wave Image Restoration Algorithm and Its Implementation by DSP,TP391.41
  16. Athermalization of Infrared Optical Sysem Based on Wavefrant Coding,TP391.41
  17. Study of Image Restoration Method Based on Improved Particle Swarm Optimization Algorithm,TP391.41
  18. Research of Image Restoration in Barcode Recognition,TP391.41
  19. The digital image based on the integration of learning technologies automatically repair theory and algorithm research,TP391.41
  20. Ultrasonic Testing of Image Restoration Techniques Based on Independent Component Analysis Theory,TP391.41
  21. Research on the Algorithm of Image Restoration and Enhancement for Removing Mist,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