Dissertation > Excellent graduate degree dissertation topics show

Despeckling of SAR Imagery Based on Markov Random Field

Author: SongZuo
Tutor: YuWenXian
School: National University of Defense Science and Technology
Course: Information and Communication Engineering
Keywords: Synthetic Aperture Radar DESPECKLING Markov Random Field Bayesian estimation Structure remains Wavelet
CLC: TP391.41
Type: Master's thesis
Year: 2007
Downloads: 121
Quote: 0
Read: Download Dissertation

Abstract


Synthetic aperture radar (SAR) is a high-resolution capability of the imaging radar range and azimuth up, round-the-clock, multi-polar, multi-frequency, multi-angle data access, and with some clouds, rain, smoke, vegetation, shallow ground penetrating ability. In recent years, the SAR has been rapid development, has been widely used in the field of battlefield detection, high-altitude camera, earth remote sensing, mapping, resource exploration, disaster prediction. Speckle reduction is the most basic SAR image processing is also one of the most important issues has been widespread concern by domestic and foreign scholars. Its core problem is how to suppress speckle while maintaining the structural characteristics of the image points, lines, edges, texture. Both spatial filtering, multi-resolution speckle, Bayesian blotches, or a variety of the latest mathematical tools speckle algorithm, not without a trade-off between the spots suppression and structure remains, often fully suppresses speckle but also excessive smooth structural features or structure to maintain good spots but residual lot, does not yet exist an algorithm can be a good solution to this contradiction. MRF (Markov Random Field, Markov random field) is an effective tool for pixel spatial correlation modeling, by virtue of his position as a priori model Bayesian speckle method is one of the hot research field speckle. Papers from the two aspects of the image domain and wavelet domain to explore new ideas and new ways to MRF for SAR image speckle, these methods, or directly from the MRF model to start, or to take measures to strengthen the structure remains outside from the model, and the fundamental purpose of spots are still efforts to solve the contradiction between the suppression and structure remains. In the image domain through the analysis of the MMRF and GMRF existing models, it is pointed out that they can not describe the structural characteristics of the image lies unreasonable weight parameter settings, this paper model both inside and outside to start strengthening the structural features remains. A whole new structure to maintain the characteristics SPMRF model, it can adaptively according to the local characteristics of the image to the right to adjust the model weight parameters, to overcome the MRF model can not effectively modeling structural features, as a priori model. Bayesian speckle ideal speckle effect; in the original fixed neighborhood MMRF model-based speckle method by introducing adaptive adjustment mechanism of the neighborhood, the proposed the AN-MMRF speckle method, both fully play to the MMRF model calculation of a small amount of speed, noise suppression can take advantage of, and to overcome the structural ambiguity caused by the original fixed neighborhood MMRF speckle structure remains on has been greatly improved. In the wavelet domain hidden Markov tree and hidden Markov random field model combining hidden state estimation method based on the of HMT-HMRF model of SAR image wavelet coefficients, the method takes full advantage of the scale and scales of the wavelet coefficients within, to improve the accuracy of the classification of the wavelet coefficients estimated to weaken noise led wavelet coefficients and maintain signal led wavelet coefficients using Bayesian laid the foundation. SAR image based on the method of wavelet speckle effectively suppress speckle while maintaining the structural characteristics of the image.

Related Dissertations

  1. Research on Electromagnetic Acoustic Transducer Detection System Based on FPGA,TH878.2
  2. Signal Detection Circuit Design for Slow-Light Optic Fiber Gyroscope,V241.5
  3. Research on Testing and Analyzing Technology for Time Parameter of Aerospace Relay,TM58
  4. The Research on Ship Imaging Algorithm under the Sea Clutter Background,TN958
  5. Modeling of Sea Clutter and Effect of Sea Clutter on Isar Imaging,TN957.54
  6. ISAR Imaging Simulation of Space Targets and Target Recognition Based on ISAR Images,TN957.52
  7. Voice Signal De-noise Base on Wavelet and Arithmetic Implement with DSP,TN912.3
  8. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  9. Research on Feature Extraction and Classification of Doppler Ultrasonic Blood Flow Signals of Wrist Radial Artery,TP391.41
  10. Study of Event Related Potentials Based on Chinese Auditory Cognition,R318.0
  11. Based on statistical methods of magnetic resonance imaging of the human brain image segmentation and three-dimensional analysis of the data,R445.2
  12. Design of Testing Equipment for Electronic Products,TN06
  13. Research on Recognition of Fabric Weave Pattern Based on Space-Frequency Domain,TS101.923
  14. Research on Identification System of Cashmere and Wool Fiber,TS101.921
  15. Measurement and Reconstruction of Ion Channel Current for Cell Membrane in Rat’s Cerebral Cortex,Q42
  16. Gas Concentration Analysis and Realized Based on Lionking Platform,TD712
  17. The Design of the Coal Mine’s Safety Evaluation System Based on Wavelet Neural Network,TD79
  18. Characteristics of sensory stimulation evoked,R318.0
  19. Network transmission ROI image coding algorithm,TN919.81
  20. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  21. Image Fusion Algorithms Based on Multi-scale Analysis,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