Dissertation > Excellent graduate degree dissertation topics show

Research on Stereo Matching Technology of Binocular Parallax Measurement

Author: SunZuo
Tutor: LiangZhongCheng
School: Nanjing University of Posts and Telecommunications
Course: Optical Engineering
Keywords: parallax triangulation measurement imaging sensor stereo matching
CLC: TP391.41
Type: Master's thesis
Year: 2013
Downloads: 75
Quote: 0
Read: Download Dissertation

Abstract


The principle for people to get the distance of surroundings is parallax ranging. For the development of optoelectronic devices, computer and software technology, stereo vision technology based on binocular parallax has been widely used at military confrontation, industrial production, robot vision, engineering prospecting and so on.This paper focus on the method of imaging registration based on the region matching algorithm.Cameras or videos image a object at two different points, so two images of the object have a bit of deviation called parallax. If know parallax, combined with related parameters of system, the depth information of the object can be obtained simply.The process of parallax ranging has five parts:obtaining images, camera calibration, image rectification, stereo matching and3D information extraction. Camera calibration is the basic of system. The accuracy of calibration will greatly influence stereo matching accuracy and depth information obtaining. Image matching is the key of parallax obtain, and restraint to parallax ranging is distinguished by different matching strategies. This paper researched current camera calibration technologies and image matching technologies, compared the characteristics, shortcoming and application area. For camera calibration, paper discussed using Matlab calibration toolbox and OpenCV camera calibration method. For image matching technology, the paper discussed using overall Graph Cut matching method and regional Blocking Matching method. Besides, experimental verification has been done for them.According to theoretical analysis, the project constructed dual-cameras ranging system, which could measure distance in30m and let relative error under2%. Then, the paper confirmed system parameters which farther upon500m and less than5%error values. At last, the paper did theoretical derivation for system distance resolution and system measuring accuracy and gives the relationship curve between system structures and them. It provides reference value to using the ranging system in practice in the future.

Related Dissertations

  1. Simulation of Underwater Scene Based on GPU,TP391.41
  2. Research on Key Technologies for Binocular Stereo Vision,TP391.41
  3. Research and Implementation of 3D Reconstruction Algorithm Based on Computer Vision,TP391.41
  4. Research on Detection System of Moving Object Based on Binocular Vision,TP391.41
  5. Research on Binocular Stereo Vision System for Indoor Mobile Robot on Complex Environment,TP242
  6. Texture Mapping Technology of a Virtual Liver Surgery,TP391.41
  7. Research on Dense Stereo Matching,TP391.41
  8. Parallax Calibration of Multi-viewpoint Images,TP391.41
  9. Investigation of Full-parallax Holographic Stereogram Using the One-step Laser Writing Method,O438.1
  10. Based on the binocular vision of 3D information acquisition and reconstruction method,TP391.41
  11. Depth Extraction and Virtual View Synthesis from Multi-view Stereo Images,TP391.41
  12. Design of Tear Bomb Intelligent Casting System Based on Binocular Stereo Vision,TJ413
  13. Research on Stereo Video Image Compression,TP391.41
  14. Multiwavelet-Based Digital Image Watermarking,TP309.7
  15. Glasses -dimensional imaging method of,TP391.41
  16. Research on the Effect Evaluation of ASDS Based on Parallax,TP391.41
  17. Visual Fatigue and Mitigation for Sterroscopic Image Observation,TP391.41
  18. The Research, Application and Implementation of Distribute Source Coding,TN911.21
  19. Research and Application of 3D Reconstruction on Moving Vehicle Based on Binocular Stereo Vision,TP391.41
  20. Stereo vision based multi-target tracking method,TP391.41
  21. Study of Depth Measurement Based on Integral Imaging,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