Dissertation > Excellent graduate degree dissertation topics show
Segmentation of Retinal Blood Vessels Based on Data-Driven Markov Chain Monte Carlo
Author: ZhouZuo
Tutor: ChenSongCan
School: Nanjing University of Aeronautics and Astronautics
Course: Applied Computer Technology
Keywords: Markov Monte Carlo Data-driven Vessel segmentation Curvelet transform Medical Image Processing
CLC: R318.0;TP391.41
Type: Master's thesis
Year: 2011
Downloads: 83
Quote: 0
Read: Download Dissertation
Abstract
|
Retinal blood vessels as the body non-invasive observation of the vital organs , which can reflect the different degrees of change in blood pressure, atherosclerosis and other cardiovascular diseases, symptoms, particularly when associated with vascular lesions occur in organs , retinal vessel diameter , curvature , etc. This change in characteristics to some extent, reflect the extent of lesions . Therefore, quantitative and qualitative analysis of retinal vascular automatically has a very important clinical value, and retinal vessel segmentation and extraction of blood vessels is the primary task . Most existing segmentation method although retinal vascular lesions of non- vascular image segmentation has good effect, but the effect of the lesion image segmentation is still not ideal , You Duiguang according uneven lesions and other sensitive . To this end , this paper presents a relatively robust vascular segmentation. This method is the first attempt to use computer vision Top-down and Bottom-up two kinds of hierarchical segmentation framework by combining the retinal image segmentation. Specifically, first, the retinal image of green color channel transform using Curvelet vessel enhancement . Then Bayesian statistical framework , using a reversible jump Markov Monte Carlo algorithm to search the parameter space , thus obtained does not depend on the initial segmentation approximate global optimal segmentation, while taking advantage of data-driven mean shift clustering algorithm and Canny edge detection operator to accelerate Markov chain dynamics . This article in the MATLAB environment, using standard STARE retinal image library four images , the experiment results show that the method is not only images but also for non-diseased lesion images are robust to segmentation , and through pattern recognition There is no free lunch theorems of the method is , in theory, the theoretical analysis .
|
Related Dissertations
- Image Fusion Algorithms Based on Multi-scale Analysis,TP391.41
- Fundus Image Segmentation Based on SVM and Template Matching,TP391.41
- SAFFRON based automated testing framework for QTP research,TP311.52
- Scrap copper smelting process control system design and implementation,TP273
- OCT fundus image layer segmentation and detection of OD Center,TP391.41
- Based on DCE-MRI of computer-aided diagnosis of benign and malignant breast lesions studied,TP391.7
- Automated Testing Framework System ATestPPMC Research and Implementation,TP311.53
- Application of Bayesian Method in Risk Based Inspection,X928.03
- Particle Filter Algorithm in Intelligent Transport System Research,TP29-AC
- Road Extraction Technolegy Based on Multi-Sensor Image Fusion,TP391.41
- Vascularity-Oriented Level Set Algorithm for Vessel Segmentation in Medical Image Processing,TP391.41
- The Study on Seismic Data Denoising and Interpolation with Curvelet Thresholding Iterative Method,P631.44
- A Bayesian Probabilistic Approach to Structural Parameter Identification and Application,O212
- The application of image processing technology in the study of neurological and psychiatric disorders,R749
- Research on the Piecewice Constant Level Set Method and the Mumford-Shah Model for Medical Image Segmentation,TP391.41
- Study on Outstanding Claims Reserve of R&C Based on Generalized Addictiv Mixed Model,F840
- Based on the second generation Curvelet transform image fusion research,TP751
- Corpus vocabulary teaching in high school experimental study,G633.41
- Automated software testing technology and applied research,TP311.52
- Based on B / S structure of the study and implementation of automated testing,TP311.52
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
|