Dissertation > Excellent graduate degree dissertation topics show

Visual Perception-based Image and Video Quality Assessment

Author: DengQinGeng
Tutor: GaoXinBo
School: Xi'an University of Electronic Science and Technology
Course: Signal and Information Processing
Keywords: Image quality Human visual perception Objective quality assessment JPEG compression Video Database
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 324
Quote: 3
Read: Download Dissertation


Objective quality assessment plays an important role in image and video applications. Image / video signal acquisition, compression, storage, communication and transmission process will suffer more or less distortion. Therefore, it is necessary to establish an objective, effective, and easy to assess the degree of distortion of the image / video or fidelity algorithm platform. In the practical application of the process, the objective evaluation can be used as a criterion or measure used to calibrate the processing system, or for algorithm optimization and parameter optimization, can also be used as a means for evaluating the performance of the algorithm. The human eye as the final receiving end of the visual information, and its perception of the image / video quality is the most accurate. Therefore, if the objective quality assessment algorithm human visual perception characteristics will improve the accuracy of the algorithm. Full reference type objective evaluation of image quality from the existing structure similarity algorithm, the error in the distortion of the image structure information, location, and quantity of information in the pixel domain fusion, and fully consider the characteristics of human visual perception. a full reference type image quality assessment method based on perception and content. Experimental results show that the correlation between the evaluation results of the algorithm and subjective scores have been greatly improved. At the same time, the algorithms used in video quality assessment, its performance is superior to the organization announced VQEG dozens of video quality evaluation model. No reference type for JPEG compressed images objective quality assessment issues, this paper presents two transform domain-based non-reference type image quality assessment algorithm. The first algorithm in the DCT domain of the image quality estimation, the algorithm needs only partially decoding the JPEG compressed stream, and by the proportion of statistical quantized zero coefficient, the joint of the human eye contrast sensitivity function to the compressed image quality assessment. The algorithm is taken directly on the compressed stream processing, greatly improves the speed of operation of the algorithm. Wavelet transform the second algorithm is the first decoded image, and then by the statistical characteristics of the change in the Wavelet domain detection coefficient to measure the quality of the distorted image. Experimental results show that PSNR performance of the algorithm is superior to both transform domain-based quality evaluation of performance without reference type quality assessment algorithm. This paper extends the the video test VQEG database. In order to evaluate the distortion produced during the compression process on the currently popular MPEG-4 and H.264 standard, and create a new video test database. All test video is compressed using the MPEG-4 and H.264 algorithm and subjective method of scoring. The establishment of the database has laid a good foundation for further study of the video quality assessment algorithm.

Related Dissertations

  1. Evaluation of Image Quality of Megavoltage Computed Tomography for Helical Tomotherapy Unit and Aplication for Dose Recalculation,R730.5
  2. Distortion effects on image quality evaluation and classification,TP391.41
  3. Image Quality Evaluation Based on HVS research,TP391.41
  4. Stationary orbit satellites Area CCD imaging simulation analysis and integration time of the feature,TP751
  5. Clinical Study of the Diagnosis of Coronary Heart Disease with Dual-source Computed Tomography,R541.4
  6. The Preliminary Study of Animal Model Built and Clinical Imaging Analysis of Crypotococcal Meningoencephalitis,R445.2
  7. The Experimental Study on Lucky Imaging Technology for Spatial Objects,TP391.41
  8. The Research on the Astronomy Imaging Technology of Diffraction-Limit Based on Lucky Imaging,TP391.41
  9. Digital image based on statistical features resampling detection method,TP391.41
  10. Spacecraft orbit and attitude parameters impact analysis for optical imaging,V412.41
  11. The Theoryetical and Numerical Study on Lucky Imaging Technology,TP391.41
  12. X-ray Grazing Incidence Telescope Image Quality Evaluation and Computer Simulation,TH743
  13. Blind Detection of Digital Images Tampering,TP391.41
  14. A Unified Objective and Subjective Image Quality Evaluation Model Based on the Human visual System,TP391.41
  15. The tampered image authentication technology content - based multi-zone,TP391.41
  16. Design of Zoom Microscope,TH742
  17. Visual attention and cognitive behavioral model and its application,TP391.41
  18. The Experimental Study on Lucky Imaging Technology,TP391.41
  19. The Application of ASMA-FAM and MSCT in Patients with Coronary Heart Disease,R541.4
  20. Virtual Scene Generating System Based on Video Materials,TP391.41
  21. Research About Tyre Nondestructive Detection System X Ray Imaging,TP274.51

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