Dissertation > Excellent graduate degree dissertation topics show

Research on Intensity Inhomogeneous Image Segmentation Based on Level Set

Author: WuZuoFeng
Tutor: HuangZuo
School: South China University of Technology
Course: Signal and Information Processing
Keywords: intensity inhomogeneity image segmentation level set method active contourmodel casting defects detection
CLC: TP391.41
Type: Master's thesis
Year: 2013
Downloads: 20
Quote: 0
Read: Download Dissertation

Abstract


With the increasing development of the computer vision technology, the auto-detectiontechnique based on image processing has become a hotspot. As a key technology, imagesegmentation is the difficulty of research. Especially，the intensity inhomogeneous imagesbring out a great challenge in image segmentation. Intensity inhomogeneous imagesegmentation is widely used. For example，in the industrial field，Image segmentationproblem with intensity inhomogeneity need to be solved in the X-ray based casting defectssmart detection algorithm research due to the intensity inhomogeneity in X-ray imaging.Recently，level set based active contour model has become a hotspot in image segmentationresearch due to the advantage of changing its topological structure to adapt to the complicatedshape of the objective，simple numerical implementation and achieving the segmentation ofobjective completely. This dissertation launches series of research about level set method withits Application in image segmentation，makes a thorough analysis about the regioninformation based active contour model and proposes several segmentation algorithmsapplied to image with complex background structure and intensity inhomogeneity. The mainresearch work of this dissertation can be summarized as follows：First，the definition and evolution of the level set function are discussed and analyzed.Then this dissertation focus on the re-initialization problem during the evolution of the levelset and presents a new distance regularized term according to the analysis of side effect aboutthe several distance regularized terms proposed，which maintain the level set function thesigned distance property in the narrowband around the zero level set. On this basis，anarrowband implementation of the level set is presented，which play an effective role in theobjective segmentation with complex background structure.Second，this dissertation introduces several region information based active contourmodels proposed and discusses the problem that the segmentation result of the several modelsapplied to intensity inhomogeneous image segmentation are all restrained by the Convolutionkernel window size and sensitive to the contour initialization. Then an improved activecontour model based on bias field correction estimated by local region information ispresented. The new model contains the original model as the local term and introduces aglobal linear fitting term as the global term，which adaptively set the coefficients of the globaland local term according to the changing degree of gray level among different regions. The improved algorithm enhances the ability to expand the target capture range and the robustnessof the segmentation results regard to contour initialization.Third，This dissertation presents a casting defects detection algorithm based on level setbased active contour model. The algorithm accomplishes the defects localization through alinear track scanning method，which contribute to extracting the defects detection region ofinterest and setting the initial contour automatically. Then an improved active contour modelbased on bias field correction estimated by local region information is presented, which ismore applied to casting defects segmentation. Finally，several criteria is presented to sift thecasting defects, which can remove false defects.Finally，This dissertation point out the Inadequacies of our algorithms and gives theforecast of the segmentation development.

Related Dissertations

  1. Research on Image Recognition Algorithm in the Forest Fire Prevention System,TP391.41
  2. Application Research of Digital Image Processing on Container Inspection,TP274.4
  3. Study on the Design and Biomechanical Behaviors of Individualized Artificial Knee Joint,R318.1
  4. Based on statistical methods of magnetic resonance imaging of the human brain image segmentation and three-dimensional analysis of the data,R445.2
  5. Road extraction algorithm based on region segmentation of remote sensing image,TP751
  6. Tobacco Diseases Auto-Recognition Research Based on Image Processing Technology,S435.72
  7. The Study on R-M Instability of the Material Interface in Gas-water Compressible Flow,O359.1
  8. Research for Infrared Image Target Identification and Tracking Technology,TP391.41
  9. BF-FCM Clustering Algorithm and Its Application in the Image Segmentation,TP391.41
  10. Study of High Resolution Image Road Extraction Method Based on Morphology Strategy,TP751
  11. Research of Color Image Segmentation Method and Its Application Based on Visual Features,TP391.41
  12. The Research of Medical Image Fusion Algorithm Based on Level Set Regions Segmentation,TP391.41
  13. Building Feature Extraction from LiDAR Point Cloud and CCD Image,P225.2
  14. A Jump Condition Capturing Scheme for Elliptic Interface Problems,O241.82
  15. Segmentation of cDNA Microarray Image Using Fuzzy C-means Algorithm Optimized by Particle Swarm,TP391.41
  16. Study on Medical Image Segmentation Method Based on Parametric Active Contour Models,TP391.41
  17. Variational multiphase image segmentation model and Split Bregman iterative algorithm,TP391.41
  18. The surface geometry noise removal of non- local variational model,TP391.41
  19. Image Segmentation and Registration for Bone Scintigraphy Images,TP391.41
  20. Image Segmentation Based on Active Contour Models,TP391.41
  21. Natural Images Segmentation Based on Active Contours Models and Saliency Map,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