Dissertation > Excellent graduate degree dissertation topics show

Research on Algorithm of Moving Object Segmentation in Video Sequences

Author: LiQingTao
Tutor: ZhangYouZhi
School: Shandong University
Course: Communication and Information System
Keywords: Moving Object Segmentation Accumulated frame difference Gradient image Moving object mask
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 99
Quote: 1
Read: Download Dissertation

Abstract


With the rapid development of information technology and computer Internet and multimedia information in people's lives, more and more applications. For multimedia information processing technology gradually developed. Multimedia information larger amount of data, particularly important to study efficient compression encoding method. As the MPEG-4 video coding standards an important part, the object-oriented video coding technology is the key to improve the coding efficiency, and an efficient video object segmentation algorithm is the basis of object-oriented coding. Video segmentation technology has a wide range of applications in the field of image processing, computer vision. The accuracy of segmentation directly affect the effectiveness of the follow-up, so it has very important significance. Segmentation of moving objects in video sequences, domestic and foreign scholars have done a lot of research work and made a lot of segmentation method, but so far no one universal, reliable automatic segmentation algorithm. Video sequences for low bit rate multimedia applications in the research and analysis of traditional classical segmentation method based on a simple and fast moving object segmentation algorithm. Firstly, using an improved method of frame difference accumulated difference image, frame image obtained adjacent frame difference and discontinuous frame difference, then these two results \differential results the cumulative obtained difference image. Canny edge detection to segment the moving object frame image obtained gradient image. Revealed in order to eliminate background due to the difference image contains not only the movement of objects, further comprising revealed a background portion, to obtain only the moving object image templates, can be the difference image and the gradient image phase \The template for the Hough transform continuous moving object boundary. Moving object mask after morphological processing and secondary scan last to fill the area with the gray values ??of the original image to obtain the segmentation result. So that full use of the correlation between each frame image in the video sequence to improve the utilization of the image, to obtain a more precise motion of the object's edges. The algorithm is a spatial and temporal combination of a moving object in a video sequence segmentation algorithm, using the edge of the object information in a time domain frame difference and airspace combined, the algorithm is simple, fast and accurate. The experiments show that the algorithm for low bit rate video broadcast video sequence can achieve better segmentation effect.

Related Dissertations

  1. Video object segmentation algorithm,TN911.73
  2. Research on Multiple Moving Objects Recognition Technology Based on Video,TP391.41
  3. Research and Application of Motion Segmentation Based on Parameter Estimation,TP391.41
  4. Research of Twin Gradient MRI Imaging Parameters and Clinicial Application,R445.2
  5. The Research of the Moving Object Segmentation Technology Based on MPEG-4,TP391.41
  6. Based on optical flow and level set method , two - phase and multiphase moving object segmentation,TP391.41
  7. Research on Moving Object Segmentation and Its Implementation in Embedded System,TP391.41
  8. Research and Implement Based on Face Recognition with Local Gradient Derivative Patterns,TP391.41
  9. Video image processing key algorithm and intelligent video surveillance system,TP391.41
  10. Research on Video Moving Object Segmentation Based on H.264 Coding Standard,TP391.41
  11. Research on Gait Object Extraction and Recognition,TP391.41
  12. Study on the Morphing of Color Facial Images Based on Improved MR-ASM,TP391.41
  13. Improved anisotropic diffusion method in image denoising,TP391.41
  14. Research on Rate Control and Scalable Coding for Video Communication,TN919.81
  15. Complex background of the video moving target tracking,TP391.41
  16. Research on Object Tracking and Classification for Intelligent Video Surveillance,TP391.41
  17. Research on the Monitoring of Discontinuous Deformation by Differential SAR Interferometry,P228.4
  18. Merging Facial Images with Visual Feature Extraction,TP391.41
  19. Based on the multi- target detection and tracking of a moving platform,TP391.41
  20. Study of Video Enhancement Algorithms for HDTV,TN941.3

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