Publications
Text queries can be conducted by Author, Title, or Keyword.
Efficient skeletonization of volumetric objects
Source: IEEE Transaction on Visualization & Computer Graphics 1999;5(3):196-209.
Author: Zhou Y, Toga AW.
Abstract:
Skeletonization promises to become a powerful tool for compact shape description, path planning, and other applications. However, current techniques can seldom efficiently process real, complicated 3D data sets, such as MRI and CT data of human organs. In this paper, we present and efficient voxel-coding based algorithm for skeletonization of 3D voxelized objects. The skeletons are interpreted as connected centerlines, consisting os sequences of medial points of consecutive clusters. These centerlines are initially extracted as paths of voxels, followed by medial point replacement, refinement, smoothness, and connection operations. The voxel-coding techniques have been proposed for each of these operations in a uniform and systematic fashion. In addition to boundary complexity, explicit extraction of ready-to-parameterize and branch-controlled skeletons, and efficient object hole detection. These issues are rarely discussed in traditional methods. A range of 3D medical MRI and CT data sets were used for testing the algorithm, demonstrating its utility. INTRODUCTION: Modern techniques enable the generation of large 3D volume data sets with high resolution, such as MRI and CT data sets. Skeletonization of such a volume, theoretically, promises a compact description of discrete objects, providing an efficient method for visualization and analysis, such as feature extraction, feature tracking, surface generation, or automatic navigation. In this paper we propose a simple, fast, and efficient skeletonization algorithm, which employs voxel-coding techniques and directly focuses on real 3D volume data. Our algorithm survives the challenge of a variety of MRI and CT data sets, characterized by connectivity preservation, centeredness satisfaction, straightforward computation, no sensitivity to object boundary complexity, smooth, fine, easy to control skeleton generation, and efficient object hole detection. APPLICATIONS AND RESULTS: Skeletonization introduces alternative shape descriptors. It promises to become a powerful tool for operations such as grouping, feature tracking, path planning, and bridging the gap between low-level and high-level representation of objects. CONCLUSIONS: In this paper, we describe an efficient skeletonization technique- voxel coding. Two types of voxel-coding have been proposed for the skeletonization of 3D complex objects. The SS-coding converts objects into a directed cluster graph, while BS-coding generates a traditional minimum field distance. The core idea of our algorithm is the SPE procedure. The SS-coding-based SPE procedure has been applied, respectively, for skeleton extraction, refinement, smoothness, and connection operations; the BS-coding is combined for medical point replacement wherever a shortest path is extracted. Our voxel-coding techniques has significant advantages over traditional methods, connectivity preservation,centeredness satisfaction, straightforward computation, no sensitivity to object boundary complexity, and smooth, fine, easy-to-control, and ready-to-parameterize skeleton generation. An additional benefit of our voxel-coding is that object holes or cycles are also easily detected without additional computational cost. Several applications of our algorithm are introduced and a range of real 2D and 3D data sets were used to test the algorithm, documenting its efficiency.
Source: IEEE Transaction on Visualization & Computer Graphics 1999;5(3):196-209.
Author: Zhou Y, Toga AW.
Abstract:
Skeletonization promises to become a powerful tool for compact shape description, path planning, and other applications. However, current techniques can seldom efficiently process real, complicated 3D data sets, such as MRI and CT data of human organs. In this paper, we present and efficient voxel-coding based algorithm for skeletonization of 3D voxelized objects. The skeletons are interpreted as connected centerlines, consisting os sequences of medial points of consecutive clusters. These centerlines are initially extracted as paths of voxels, followed by medial point replacement, refinement, smoothness, and connection operations. The voxel-coding techniques have been proposed for each of these operations in a uniform and systematic fashion. In addition to boundary complexity, explicit extraction of ready-to-parameterize and branch-controlled skeletons, and efficient object hole detection. These issues are rarely discussed in traditional methods. A range of 3D medical MRI and CT data sets were used for testing the algorithm, demonstrating its utility. INTRODUCTION: Modern techniques enable the generation of large 3D volume data sets with high resolution, such as MRI and CT data sets. Skeletonization of such a volume, theoretically, promises a compact description of discrete objects, providing an efficient method for visualization and analysis, such as feature extraction, feature tracking, surface generation, or automatic navigation. In this paper we propose a simple, fast, and efficient skeletonization algorithm, which employs voxel-coding techniques and directly focuses on real 3D volume data. Our algorithm survives the challenge of a variety of MRI and CT data sets, characterized by connectivity preservation, centeredness satisfaction, straightforward computation, no sensitivity to object boundary complexity, smooth, fine, easy to control skeleton generation, and efficient object hole detection. APPLICATIONS AND RESULTS: Skeletonization introduces alternative shape descriptors. It promises to become a powerful tool for operations such as grouping, feature tracking, path planning, and bridging the gap between low-level and high-level representation of objects. CONCLUSIONS: In this paper, we describe an efficient skeletonization technique- voxel coding. Two types of voxel-coding have been proposed for the skeletonization of 3D complex objects. The SS-coding converts objects into a directed cluster graph, while BS-coding generates a traditional minimum field distance. The core idea of our algorithm is the SPE procedure. The SS-coding-based SPE procedure has been applied, respectively, for skeleton extraction, refinement, smoothness, and connection operations; the BS-coding is combined for medical point replacement wherever a shortest path is extracted. Our voxel-coding techniques has significant advantages over traditional methods, connectivity preservation,centeredness satisfaction, straightforward computation, no sensitivity to object boundary complexity, and smooth, fine, easy-to-control, and ready-to-parameterize skeleton generation. An additional benefit of our voxel-coding is that object holes or cycles are also easily detected without additional computational cost. Several applications of our algorithm are introduced and a range of real 2D and 3D data sets were used to test the algorithm, documenting its efficiency.