Dissertation > Excellent graduate degree dissertation topics show

The Application and Research on Digital Image Inpainting Technology in Processing of Damaged Pictures

Author: ZhangHuiJuan
Tutor: LiLingYuan
School: Central China Normal University
Course: Communication and Information System
Keywords: Digital image restoration Texture Synthesis Template size Adaptive selection Repair priority Repair order Incremental weights operator
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 256
Quote: 1
Read: Download Dissertation

Abstract


As a very important branch of digital image processing, digital image restoration techniques are widely used in the damaged pictures, photographs, paintings, and repair of movie film; repair of ancient relics; removal of the excess of the target, text or objects; production of film and television special effects; transmission and compression of the image; image of the Zoom In and other fields. So far, the research and application of digital image restoration techniques are divided into variational PDE model-based digital image restoration and digital image-based texture synthesis repair the two technologies. Digital image restoration techniques based on variational PDE model through mathematical method to establish partial differential equations the defect border area of ??the image to determine the illumination direction using diffusion theory, valid information from the boundary to the defect area spread . Based texture synthesis of digital image restoration techniques, \. Currently, such technology \A.Criminisi algorithm is a more classic \search within to find the current pixel block is most similar to the mold plate to replace the current pixel block, the final completion of the repair of the defect area. A.Criminisi algorithm compared with other repair algorithm faster repair, repair, better quality, in particular when repair a defect image of the large area, and its advantages are even more prominent. , A.Criminisi algorithm has been widely used in image repair, video repair. However, A.Criminisi algorithm also has the disadvantage of template size is fixed, the time it takes to search for a matching block long, error-prone repair fill the order, the joints between blocks obvious traces. The article focuses study A.Criminisi algorithm, two improvements and shortcomings: first, to repair the defect image using A.Criminisi algorithm, regardless of whether the image information rich, the template block size always fixed not change, this For certain complex image processing, the effect is not good. In this paper, the number of gray levels of the statistical image to determine whether the information to be repaired image rich rich according to the image information or not, the template size adaptive selection, if the image information, select the smaller the template, if the image The information is not rich, then choose the larger template. Second using A.Criminisi algorithm to repair the defect image, repairing sequence in accordance with the restoration priority formula, wherein structural information amount D (p) the weightings value is not supplied with the repair process change is changed, whereby determine the repair order is not entirely rational, error-prone repair of the defect area image effect is not very good. By introducing incremental weights operator λ enhanced repair late structure the amount of information the role of D (p) to determine the priority of the repair of the defect area boundary pixels, resulting in better, more reasonable repair order, to the repair amplitude larger area of ??the damaged image obtained when better repair effect. Meanwhile, the improved algorithm proposed by two points were a large number of simulation, to verify the feasibility and effectiveness of the improved method.

Related Dissertations

  1. Video Coding Research Based on Texture Characteristics,TP391.41
  2. The Research on Texture Synthesis Technology from Cloud Theory & Been Evolution Genetic Algorithm,TP391.41
  3. Texture Synthesis Algorithm Based on Samples,TP391.41
  4. Image restoration of key technologies,TP391.41
  5. Video image segmentation based texture synthesis technology research,TP391.41
  6. Research and Application on User-Controllable Multi-Exemplars Texture Synthesis,TP391.41
  7. Adaptive Waveform Selection Techniques for Target Tracking,TN953
  8. Reliability Analysis of Three-State Standby Repairable System with Delay Repair,O213.2
  9. The nonparametric qualified support domain blind image restoration algorithm,TP391.41
  10. Interactive Volume Modeling and Illustration of Muscle,TP391.41
  11. Implementation of Transductive Support Vector Machine in Data Prediction,TP274
  12. Real-time Rendering Based on GPU for Large-scale Virtual Environment,TP391.41
  13. Research on Texture Synthesis Algorithm from Samples,TP391.41
  14. Texture Synthesis and Classification Based on Support Vector Machines and Ensemble Learning,TP391.41
  15. Research on Digital Image Completion Methods and Its Application in Image Compression,TP391.41
  16. Research of Runtime Synthesis Large-Scale Vegetation Landscape Based on the Resource Interaction,TP391.41
  17. Digital image restoration algorithm,TP391.41
  18. Video Coding System Based on Texture Analysis and Synthesis,TN919.81
  19. Research on Algorithms for Near-Optimal Detection for V-BLAST Systems,TN929.5
  20. Urban Road Intersections bus priority signal control technology,U491.54
  21. Texture generation mapping technology research and application,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