Dissertation > Excellent graduate degree dissertation topics show

Video Segmentation Algorithm Design Based on the Background Reconstruction

Author: MiJuanFang
Tutor: JiXiaoPing
School: Taiyuan University of Technology
Course: Communication and Information System
Keywords: Video objects segmentation Frame differential Background subtraction Background reconstruction
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 28
Quote: 0
Read: Download Dissertation

Abstract


Video also called dynamic images, the activity image or image motion,is a group of images in time the shaft. Not only it contains a static image contains contents in the scene, but also contains target motion information and the object Video objects segmentation is always one of the difficult and hot research at home and abroad in video monitoring field, the purpose of which is to extract video objects from video scene which people are interested with or something have important properties of one or more. And the region which people are interested in usually has movement characteristic, so video division normally refers to the moving object detection and tracking. The effective segmentation of moving targets is vital to target identification, classification, trace etc. late operation.Background Difference value method is a fast and accurately method getting moving targets,but it’s tend to be impacted by the environment like sunshine,which results in a not very good measuring result. So how to get the real-time background image will become the difficulty and the focus. This paper gives a new method of background reconstruction algorithm based on the shortage of various reconstruction algorithm, taking a good effect.This paper is to use background difference method along with the method of analytic frame differential, difference lies in adopting three frames difference-method to detect the change region, after getting the moving targets, completing background image by reconstructing the background. Background-reconstruction is by way of accumulative atlas mapping through change detection results,constantly replacing initial pixels with a new background pixels, finally getting relatively complete update background image.At last through background difference method to get more ideal target motion. Experiment shows that the algorithm can better remove background occluded area, effectively realize video object division, get a accurate and complete moving targets, and the method is very easy to realize, have good segmentation stability.

Related Dissertations

  1. Intelligent video detection technology based on the traffic lights control,TM923.5
  2. Vision-based Human Movement Analysis,TP391.41
  3. Video surveillance system moving target detection algorithm,TP391.41
  4. Research on Moving Objects Detection Algorithms in Security Monitoring System,TP391.41
  5. Research & Implementation of Video Vehicle Detection and Tracking System,TP29
  6. The Research of Gait Recognition Based on the Figure and Force-Field Transformation,TP391.4
  7. Moving Objects Detection in Complex Background,TP391.41
  8. Research on Moving Object Detection and Tracking in Sequence Image,TP391.41
  9. Study of Methods for Moving Human Detection and Tracking in Image Sequences,TP391.41
  10. Research of Motion Detection and Tracking Algorithm in Video Stream,TP391.41
  11. Theoretical and Experimental Study on Effective Extraction of Glucose Specific Signal in NIR Spectroscopy,TH776
  12. Research on Moving Objects Detection in Video-based Surveillance System,TP277
  13. Movement Goal Examination and Video Monitoring Software Development,TP391.41
  14. Research on Target Tracking Technology of Digital Monitoring System,TP391.41
  15. Research on Shadow Detection Techniques for Moving Vehicles in Video Sequences,TP391.41
  16. The Research on Video-Based Vehicle Detection Technique for Traffic Crossing,TP274.4
  17. Study on the Intelligent Recognition of Video Surveillance in the City Tunnel,TP277
  18. The Technology Research of Moving Target Detection and Tracking,TP391.41
  19. Method and Key Technique Research about Vision-based Traffic Flow Detection,TP391.41
  20. Complex background weak target detection and tracking of,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