Dissertation > Excellent graduate degree dissertation topics show

Study on Dense Stereo Image Matching Based on Parallel Computing

Author: WuZuoCui
Tutor: WeiChiMing
School: Huazhong University of Science and Technology
Course: Biomedical Engineering
Keywords: Accelerating Parallel Computing CUDA Muti-thread Dense Matching
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 100
Quote: 0
Read: Download Dissertation

Abstract


Stereo dense matching is very important in high resolution registration and 3D reconstruction. The dense matching method based on SIFT can obtain uniform and dense point cloud for object surface, which can help to obtain high resolution results, but it has low computing efficiency. Most of the 3D reconstruction and registration need to be real time, so accelerating the algorithm with parallel computing is meaningful. For designing parallel algorithm, how to choose appropriate technology according to the character of both the algorithm and device is a difficult problem. In recent years, GPGPU becomes a hot topic and has its own advantages and disadvantages compared with muti-thread. CUDA released by Nvidia in 2007 brought GPGPU into a wider field, but its performance and methods for optimization need to be researched farther.This thesis divides the stereo dense matching into several tasks and analyses the parallel and time consuming parts. Feature vector matching and dense matching parts are then designed into parallel programs with both CUDA and muti-thread. In the experiments, the performance in accelerating of these two technologies are compared and analyzed. For the CUDA based vector matching program, four different schemes are designed when task distribution for each thread is different. The performance of these schemes is compared in experiments. Also, the effect of several main optimization technologies in CUDA on the performance of program are tested in the experiments for CUDA based dense matching program.From the experiments, the computing efficiency of the parallel computing is far higher than signal computing. The technology and scheme need to be selected according to the specific algorithm and the practical hardware condition. The biggest difficulty for CUDA programming is the task distribution and the technology to optimize the communication between host and device or in device. This paper successes in accelerating the stereo matching algorithm to meet the requirements, provides theoretical reference for how to select appropriate technology when parallel program is designed for specific signal program. Also, it summarizes some useful experience for application of CUDA.

Related Dissertations

  1. The Airborne LiDAR and LiDAR Points-cloud’s Quick Processing Method,TN959.73
  2. Research on Image Digital Watermarking Technology Based on CUDA,TP309.7
  3. Application of CUDA in the Very Short Term Load Forecasting of Multi-node,TM715
  4. Application Research of GPU in Vehicle Detection and Tracking System,TP391.41
  5. Acceleration of X-ray Computed Tomography Reconstruction for Rice Tiller,TP391.41
  6. Research of Finite Element Method on GPU,O241.82
  7. Numerical Simulation of Radiofrequency Waves in Magnetized Plasma,TL612
  8. Energy Minimization Based Segmentation and 3-D Visualization for Abdominal CT Image,TP391.41
  9. Research and Implementation of the Gene Bayesian Network Construction Algorithm Based on Multi-core Environment,Q75
  10. Research on Parallel Method for Image Denoising Via Sparse Representations,TP391.41
  11. CUDA-based regular expression matching system design and implementation,TP311.52
  12. Fault Tolerance for MapReduce in the Cloud Environment,TP302.8
  13. White Light Interferometry for Fast Areal Surface Measurement Based on GPGPU,O439
  14. Research groups and evacuation simulation algorithm simulation system,TP391.9
  15. GPU-based book recommendation system Research and Implementation,TP391.3
  16. CUDA-based video fire detection system,TP391.41
  17. GPU-based reconstruction algorithm for X-rays accelerate research,TP391.41
  18. GPU-based acceleration of neutral gas leak and rescue simulation study,TP391.41
  19. Research of Image Processing Technology for Structured Light Stripe Image,TP391.41
  20. Large-scale approximation paragraph fingerprint - based page detection algorithm research,TP393.092
  21. Real-Time Image Mosaic Based on CUDA,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