Dissertation > Excellent graduate degree dissertation topics show

Research of Denosing Algorithm Especial for Video Encoder

Author: XieHong
Tutor: XuCheng
School: Hunan University
Course: Computer Science and Technology
Keywords: Filter factor Transform domain Wiener filtering Time direction of the transform domain filtering Absolute error and Motion vector
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 56
Quote: 1
Read: Download Dissertation

Abstract


Video sequence denoising problem in recent decades has been widespread concern and discussion . In real-time video surveillance , real-time streaming media services , mobile videophone applications must find , in order to take into account the image quality and real-time video denoising method is relatively simple and efficient tightly coupled with the encoder . This paper first analyzes and summarizes the existing video sequence denoising method . On this basis , the noise reduction algorithm transform domain Wiener filter and transform domain temporal filtering encoder tightly coupled experimental comparison and analysis , and pointed out the problem when they run low configuration T264 encoder . In this paper, the use of the strong correlation between the 4 × 4 blocks within the data in the calculation of the DCT coefficients normalized covariance matrix for the transform domain Wiener filtering algorithm to noise performance sharp decline in the the low configuration T264 coding environment , Apart from the DC coefficient , the variance of the remaining AC coefficients are equal to the first AC coefficient of variance . Transform domain Wiener filter in the experiments show that in the known noise strength and low profile T264 coding environment for low slow and relatively flat screen of the video sequence , the improved compression efficiency , the average peak signal to noise ratio on the two performance evaluation index . superior to several other algorithms , and still has a relatively stable in the case of the noise intensity noise reduction capabilities . Time direction for transform domain filtering in the low configuration T264 coding environment , the filter factor over - concentrated in the vicinity of the threshold of this issue , this paper presents a classification applicable to H.264 nonlinear filtering factor is determined program that is based on the absolute error and the motion vector selection with suitable filtering factor obtained by the experience , and is updated in units of blocks . The experiments show that in the known noise strength and low profile T264 coding environment , improved transform domain time-direction filtering by manually adjusting the classification number and the associated threshold value , and relatively good noise reduction effect can be obtained , and improved filter factor determining program without estimated noise variance , the computational complexity is lower than the original algorithm .

Related Dissertations

  1. Moving target trajectory analysis based Intelligent Traffic Monitoring System,TP277
  2. Array AOI system error correction algorithm design and implementation,TP274
  3. Projection-based digital video stabilization technology,TP391.41
  4. Drift compensation based on motion vector and H.264 video watermarking algorithm,TP309.7
  5. Research on Dual Information Source Model-Based Collaborative Filtering Algorithms,TP301.6
  6. Motion Object Detection Based on MPEG-4 Video,TP391.41
  7. H.264-based Digital Video Watermarking Scheme,TN919.81
  8. Real-time feedback based on user collaborative filtering algorithm,TP301.6
  9. Electronic Image Stabilizing Technology Research in Omni-directional Camera System,TP391.41
  10. Research on Anti-Noise Matching Criteria of Motion Estimation,TP391.41
  11. Based on MPEG-2 video watermarking technology research,TP309.7
  12. Research and Implementation of MPEG-4 video watermarking technology,TP309
  13. Design and Implementation of a Temporal Error Concealment Algorithm Based on Motion Vector Recovery,TN919.81
  14. Based transcoding digital video watermarking research,TP309.7
  15. Video sequences of electronic image stabilization technology,TP391.41
  16. Digital Video Watermarking Algorithm Based on Motion Vectors for AVS,TP391.41
  17. Global Motion Estimation Algorithm Based on Motion Vector Reliability Analysis for Video,TP391.41
  18. Otimization on Motion Estimation Algorithm Based on H.264/AVC,TN919.81
  19. Video Compression Technology for Visual IP Telephony’s System,TN919.81
  20. The Optimization in Pixel Domain Video Transcoding from MPEG-2 to H.264,TN919.81
  21. Video source based on motion vector algorithm forensics,TN919.81

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