Watershed-based Weighted Graph Partitioning Active Contours

Shaorong Wang,Yisong Chen,Sheng Li,Guoping Wang

DOI: https://doi.org/10.1109/cisp.2010.5646377

2010-01-01

Abstract:A novel Weighted Graph Partitioning Active Contours method based on weighted dissimilarity is introduced. This method is easy to be extended by defining different types of similarities. And it has been greatly accelerated by a watershed pre-segmentation.

What problem does this paper attempt to address?

Decompose Image into Meaningful Regions Based on Contour Detector and Watershed Algorithm

Sheng Luo,Jing-Hua Xu,Shu-You Zhang

DOI: https://doi.org/10.3233/jifs-16653

2017-01-01

Journal of Intelligent & Fuzzy Systems

Abstract:In spite of a large volume of segmentation algorithms, it is difficult to coherently segment a series of images from the same scene. Semantic segmentation might be a solution, but it is sophisticated with heavy computing cost. Many contour detectors could locate objects accurately, but the contours are broken, and Watershed algorithm could find the subtle ridges that could connect the broken contours, but often fails for over-segmentation. We try to combine these two methods to decompose images into meaningful regions, and propose an index S-measure to measure the segmentation consistency between images. Two steps are involved: detect edges by contour detector, and fuse the broken edges by Watershed algorithm. To fuse the broken edges, two ways are proposed, dilating edges with same template and with adaptive width templates. Experiments illustrate that the decomposing is fast, effectively, meaningful, robustly, and coherently, and the segmentation is consistent between images even with some transformations.
Mesh Decomposition With Cross-Boundary Brushes

Youyi Zheng,Chiew-Lan Tai

DOI: https://doi.org/10.1111/j.1467-8659.2009.01622.x

IF: 2.5

2010-01-01

Computer Graphics Forum

Abstract:We present a new intuitive UI, which we call cross-boundary brushes, for interactive mesh decomposition. The user roughly draws one or more strokes across a desired cut and our system automatically returns a best cut running through all the strokes. By the different natures of part components (i.e., semantic parts) and patch components (i.e., flatter surface patches) in general models, we design two corresponding brushes: part-brush and patch-brush. These two types of brushes share a common user interface, enabling easy switch between them. The part-brush executes a cut along an isoline of a harmonic field driven by the user-specified strokes. We show that the inherent smoothness of the harmonic field together with a carefully designed isoline selection scheme lead to segmentation results that are insensitive to noise, pose, tessellation and variation in user's strokes. Our patch-brush uses a novel facet-based surface metric that alleviates sensitivity to noise and fine details common in region-growing algorithms. Extensive experimental results demonstrate that our cutting tools can produce user-desired segmentations for a wide variety of models even with single strokes. We also show that our tools outperform the state-of-art interactive segmentation tools in terms of ease of use and segmentation quality.
An experimental comparison of superpixels detection methods for contour detection

Xuan-Yin Wang,Chang-Wei Wu,Ke Xiang,Sen-Wei Xiang,Wen Chen

DOI: https://doi.org/10.1007/s00138-018-0927-x

2018-01-01

Abstract:Recently, many superpixels detection methods have been proposed and used in various applications. We are interested in which method is more suitable for the application of contour detection. In this paper, superpixels are evaluated on BSDS500 dataset in two different aspects. On the one hand, contours are directly provided by the boundaries of superpixels and experiments show that better results could be achieved by the superpixels with irregular shapes than those with regular shapes and similar sizes. On the other hand, contours are further detected from those candidate positions which are confirmed by the boundaries of superpixels through the operation of dilation. In this situation, experiments show that competitive results could also be achieved by some superpixels with regular shapes and similar sizes. Besides, we propose a superpixels detection method called watershed-based graph (WG), by which superpixels with irregular shapes could be produced. Firstly, a graph is constructed from an over-segmented map which is achieved through a watershed algorithm. Then, to get the desired superpixels, the graph is segmented by merging neighbor segments in an order of decreasing similarity. Experiments show that higher efficiency could be achieved by WG with a moderate worse contour quality than its original graph-based method.
IMAGE SEGMENTATION: A WATERSHED TRANSFORMATION ALGORITHM

L. Belaid,Walid Mourou

DOI: https://doi.org/10.5566/IAS.V28.P93-102

2011-05-03

Abstract:The goal of this work is to present a new method for image segmentation using mathematicalmorphology. The approach used is based on the watershed transformation. In order to avoid an oversegmentation, we propose to adapt the topological gradient method. The watershed transformation combined with a fast algorithm based on the topological gradient approach gives good results. The numerical tests obtained illustrate the efficiency of our approach for image segmentation.

Mathematics,Computer Science
Adaptive Decomposition and Shared Weight Volumetric Transformer Blocks for Efficient Patch-Free 3D Medical Image Segmentation.

Hongyi Wang,Yingying Xu,Qingqing Chen,Ruofeng Tong,Yen-Wei Chen,Hongjie Hu,Lanfen Lin

DOI: https://doi.org/10.1109/jbhi.2023.3305830

IF: 7.7

2023-01-01

IEEE Journal of Biomedical and Health Informatics

Abstract:High resolution (HR) 3D medical image segmentation is vital for an accurate diagnosis. However, in the field of medical imaging, it is still a challenging task to achieve a high segmentation performance with cost-effective and feasible computation resources. Previous methods commonly use patch-sampling to reduce the input size, but this inevitably harms the global context and decreases the model's performance. In recent years, a few patch-free strategies have been presented to deal with this issue, but either they have limited performance due to their over-simplified model structures or they follow a complicated training process. In this study, to effectively address these issues, we present Adaptive Decomposition (A-Decomp) and Shared Weight Volumetric Transformer Blocks (SW-VTB). A-Decomp can adaptively decompose features and reduce their spatial size, which greatly lowers GPU memory consumption. SW-VTB is able to capture long-range dependencies at a low cost with its lightweight design and cross-scale weight-sharing mechanism. Our proposed cross-scale weight-sharing approach enhances the network's ability to capture scale-invariant core semantic information in addition to reducing parameter numbers. By combining these two designs together, we present a novel patch-free segmentation framework named VolumeFormer. Experimental results on two datasets show that VolumeFormer outperforms existing patch-based and patch-free methods with a comparatively fast inference speed and relatively compact design.
Image Segmentation Based on Watershed Transform

YANG Wen-ming,CHEN Guo-bin,SHEN Ye-hu,LIU Ji-lin

DOI: https://doi.org/10.3785/j.issn.1008-973x.2006.09.007

2006-01-01

Abstract:Aimed at resolving the problems of sensitivity to noise and over-segmentation existing in traditional watershed algorithm,a watershed segmentation scheme based on double opening and closing reconstruction and non-linear processing was proposed.According to the differences of scale and amplitude distribution between noise and signal in the scheme,morphological opening and closing prereconstruction of the original image was utilized by integrating with characteristics of morphological opening and closing operation,and morphological opening and closing post-reconstruction was performed after the morphological gradient's calculation.After a certain threshold transform was employed on the gradient,a nonlinear classification based on certain scale grade was proposed.An improved watershed labelling algorithm was put forward and applied based on pixel's connexity,and the concrete implementation flow was given.Simulated experimental results show that the scheme can restrain over-segmentation existing in traditional algorithms,and exactly locate edges.
Efficiently Solving the Piecewise Constant Mumford-Shah Model using Graph Cuts

Xiang Zeng,Wei Chen,Qunsheng Peng

2005-01-01

Abstract:This paper presents an efficient algorithm for solving one sort of region-based active contour models, i.e. the active contours without edges model [1]. We develop the link between traditional level-set based segmentation models and graph based discrete formulation by converting the original PDE based energy function to its discrete form, and show that it can be solved via graph cuts optimization conveniently. We also propose an efficient algorithm to regulate the smoothness of the interfaces, which plays a similar roll as the re-initialization of the signed distance function in solving level sets. Our new algorithm outperforms the traditional PDE driven level set methods in several ways. Firstly, it is significantly faster by taking advantage of the dynamic graph cuts algorithm. Secondly, it is numerically stable and it avoids some original drawbacks in solving PDEs. Furthermore, the new algorithm is less sensitive to the initialization. The experimental results show that the performance improvement is quite significant, which makes it plausible for real-time processing.
Neighborhood Aided Implicit Active Contours

Huafeng Liu,Yunmei Chen,Wufan Chen

DOI: https://doi.org/10.1109/CVPR.2006.205

2006-01-01

Abstract:We have developed a geometric deformable model that employs neighborhood influence to achieve robust segmentation for noisy and broken edges. The fundamental power of this strategy rests with the explicitly combination of regional inter-point constraints, image forces, and a priori boundary information for each geometric contour point within its adaptively determined local influence domain. This formulation thus naturally unifies the essences of the geometric and parametric snakes through automatic local scale selection, and exhibits their respective fundamental strengths of allowing stable boundary detection when the edge information is weak and possibly discontinuous, while maintaining the abilities to handle topological changes during front evolution. In particular, this paper presents an implementation of the method through local integration of the level set function and the image/prior-driven evolution forces, where the resulting partial differential equation is solved numerically using standard finite difference method. Experimental results on synthetic and real images demonstrate its superior performance.
Watersheds Algorithm for Relative Potential Energy

ZHANG Li-li,YAO Yong,LIU Zhi-jing

DOI: https://doi.org/10.3969/j.issn.1001-2400.2008.01.023

2008-01-01

Abstract:A watersheds algorithm for relative potential energy is presented.The presented algorithm calculates the relative potential energy of watersheds and compares them with the preestablished thresholds to remove the unnecessary details and get the important contour information.This method is introduced on the basis of the fast watersheds model.As a result,over-segmentation is effectively restrained,region fusing is avoided and the efficiency improved accordingly.Experimental results have shown that,compared with the original region-fusing algorithm,the efficiency has increased by 14.3%,and that the over-segmentation phenomenon has been obviously improved.
Voronoi-based Topological Contour Filling Approach for Environmental Impact Assessment

WU Pei-ning,TAN Jian-rong,LIU Zhen-yu,FENG Yi-xiong

DOI: https://doi.org/10.3785/j.issn.1008-973x.2009.02.023

2009-01-01

Abstract:A fast algorithm of color fill between contours based on the topological relation bintree of contours was proposed to achieve the real-time visualization of atmospheric environmental impact assessment(EIA).After closing the non-closed contours,a new topological relation identifying algorithm was presented by combining the Voronoi-based topology judge algorithm for natural closed contour(NCL) and the topology judge approach for boundary closed contour(BCL).Based on this new approach,the topological relation bintree of contours can be well generated for the topological contour filling application.An application in the GIS-based visualization for atmospheric EIA was illustrated.The color filling effect of the atmospheric diffusion contours shows that the proposed method has satisfactory results.
An Improved Contour Detection Method Based on Level Set and Watershed Transform

Shuang Wang,Shenggao Fu,Licheng Jiao,Xiaojing Zhang

DOI: https://doi.org/10.1117/12.832902

2009-01-01

Abstract:In this paper, an improved contour detection method based on level set and watershed transform is proposed. It is performed on coarse-to-fine approach: 1) primary contours detected by using level set evolution; 2) accurate contours detected by watershed transform. The combined method utilizes the advantages of level set method and watershed transform. Experimental results show that our method could detect accurate contours and gain great improvement compared with level set evolution and watershed transform.
Image segmentation method based on improved graph cut algorithm

B. Liu,Haibin Li,Y. Du,Yuxian Li,Yaqian Li,Qiang Zhang

DOI: https://doi.org/10.1109/CSAE.2012.6272787

2012-05-25

Abstract:Image segmentation method based on improved graph cut algorithm is proposed in this paper to improve the speed of graph cut and solve the watershed transformation over-segmentation phenomena. First, the image is segmented into a large number of small partitions making use of the watershed transformation. Then, a simplified network graph is constructed using the partition information. Finally, the graph cut algorithm is employed to realize network graph segmentation and get the segmentation results. The experimental results indicate that the proposed algorithm not only solves the watershed transformation over-segmentation phenomena, but also improves the speed of the graph cut algorithm.

Computer Science
Joint Dense-Point Representation for Contour-Aware Graph Segmentation

Kit Mills Bransby,Greg Slabaugh,Christos Bourantas,Qianni Zhang

DOI: https://doi.org/10.1007/978-3-031-43898-1_50

2023-06-21

Abstract:We present a novel methodology that combines graph and dense segmentation techniques by jointly learning both point and pixel contour representations, thereby leveraging the benefits of each approach. This addresses deficiencies in typical graph segmentation methods where misaligned objectives restrict the network from learning discriminative vertex and contour features. Our joint learning strategy allows for rich and diverse semantic features to be encoded, while alleviating common contour stability issues in dense-based approaches, where pixel-level objectives can lead to anatomically implausible topologies. In addition, we identify scenarios where correct predictions that fall on the contour boundary are penalised and address this with a novel hybrid contour distance loss. Our approach is validated on several Chest X-ray datasets, demonstrating clear improvements in segmentation stability and accuracy against a variety of dense- and point-based methods. Our source code is freely available at: <a class="link-external link-http" href="http://www.github.com/kitbransby/Joint_Graph_Segmentation" rel="external noopener nofollow">this http URL</a>

Computer Vision and Pattern Recognition,Machine Learning,Image and Video Processing
Gushing And Immersion Alternative Watershed Algorithm

Tianzhou Chen

2001-01-01

Abstract:In this paper, a fast and flexible watershed algorithm in digital grayscale images is introduced, which imitating the gushing and immersion process. The algorithm is accuracy and less consuming in time and memory space. It can easily get the information of the regions, and can remove the noise during segmentation.
Watershed-based Textural Image Segmentation

Shuang Wang,Xiuli Ma,Xiangrong Zhang,Licheng Jiao

DOI: https://doi.org/10.1109/ispacs.2007.4445886

2007-01-01

Abstract:The watershed transform is a well-established tool for image segmentation. However, watershed segmentation is often not effective for textural images. In this paper, we describe an improved watershed segmentation algorithm combined with texture features. The aim of this study is to improve the generalization of watershed techniques and to construct a well segmentation of textural images. The method includes two stages. The first stage is standard watershed algorithm. The second stage is processed by a clustering algorithm, fuzzy c-means (FCM). Watershed algorithm provides small homogenous patches which are merged by clustering algorithm based on texture features. The experimental results demonstrate that the combined algorithm is effective for textural image segmentation.
A Fast Watersheds Approach For Image Segmentation Based On Relative Potential Energy

Lili Zhang,Xiaojun Zhang,Zhijing Liu

2007-01-01

Abstract:Traditional watersheds methods face the difficulty of "over-segmentation" for images with many gray details and noises. In this paper, the traditional segmentation method is improved. Firstly, "Opening and Closing in morphology" is used to smooth image, and some unnecessary details are removed. Then, this paper uses the fast watersheds approach for image segmentation based on relative potential energy method. The original potential energy is replaced by relative potential energy. This approach gains higher efficiency and less run time. The experiments show a remarkable reduction of segmentation regions. This approach show an improvement on the segmentation performance and it can remove the unimportant details and noise in the image to extract the clearly whole figure of the image fleetly.
Contour-enhanced Resampling of 3D Point Clouds Via Graphs

Siheng Chen,Dong Tian,Chen Feng,Anthony Vetro,Jelena Kovacevic

DOI: https://doi.org/10.1109/icassp.2017.7952695

2017-01-01

Abstract:To reduce storage and computational cost for processing and visualizing large-scale 3D point clouds, an efficient resampling strategy is needed to select a representative subset of 3D points that can preserve contours in the original 3D point cloud. We tackle this problem by using graph-based techniques as graphs can represent underlying surfaces and lend themselves well to efficient computation. We first construct a general graph for a 3D point cloud and then propose a graph-based metric to quantify the contour information via high-pass graph filtering. Finally, we obtain an optimal resampling distribution that preserves the contour information by solving an optimization problem. When browsing, the proposed graph-based resampling performs better than uniform resampling both for toy point clouds as well as real large-scale point clouds. Furthermore, as neither mesh construction nor surface normal calculation is involved, the proposed graph-based method is computationally more efficient than the mesh-based methods.
Weighted Laplacian Method and Its Theoretical Applications

XU Shi-jie,FANG Jia-yan,LI Xiang-yang

DOI: https://doi.org/10.1088/1757-899x/768/7/072032

2020-01-01

IOP Conference Series Materials Science and Engineering

Abstract:In this paper, we develop a novel weighted Laplacian method, which is partially inspired by the theory of graph Laplacian, to study recent popular graph problems, such as multilevel graph partitioning and balanced minimum cut problem, in a more convenient manner. Since the weighted Laplacian strategy inherits the virtues of spectral methods, graph algorithms designed using weighted Laplacian will necessarily possess more robust theoretical guarantees for algorithmic performances, comparing with those existing algorithms that are heuristically proposed. In order to illustrate its powerful utility both in theory and in practice, we also present two effective applications of our weighted Laplacian method to multilevel graph partitioning and balanced minimum cut problem, respectively. By means of variational methods and theory of partial differential equations (PDEs), we have established the equivalence relations among the weighted cut problem, balanced minimum cut problem and the initial clustering problem that arises in the middle stage of graph partitioning algorithms under a multilevel structure. These equivalence relations can indeed provide solid theoretical support for algorithms based on our proposed weighted Laplacian strategy. Moreover, from the perspective of the application to the balanced minimum cut problem, weighted Laplacian can make it possible for research of numerical solutions of PDEs to be a powerful tool for the algorithmic study of graph problems. Experimental results also indicate that the algorithm embedded with our strategy indeed outperforms other existing graph algorithms, especially in terms of accuracy, thus verifying the efficacy of the proposed weighted Laplacian.
A Fuzzy Geometric Active Contour Method for Image Segmentation.

Danyi Li,Weifeng Li,Qingmin Liao

DOI: https://doi.org/10.1587/transinf.e96.d.2107

2013-01-01

IEICE Transactions on Information and Systems

Abstract:In this paper, we propose a hybrid fuzzy geometric active contour method, which embeds the spatial fuzzy clustering into the evolution of geometric active contour. In every iteration, the evolving curve works as a spatial constraint on the fuzzy clustering, and the clustering result is utilized to construct the fuzzy region force. On one hand, the fuzzy region force provides a powerful capability to avoid the leakages at weak boundaries and enhances the robustness to various noises. On the other hand, the local information obtained from the gradient feature map contributes to locating the object boundaries accurately and improves the performance on the images with heterogeneous foreground or background. Experimental results on synthetic and real images have shown that our model can precisely extract object boundaries and perform better than the existing representative hybrid active contour approaches.
Distributed Augmentation, Hypersweeps, and Branch Decomposition of Contour Trees for Scientific Exploration

Mingzhe Li,Hamish Carr,Oliver Rübel,Bei Wang,Gunther H. Weber

DOI: https://doi.org/10.1109/TVCG.2024.3456322

2024-09-30

Abstract:Contour trees describe the topology of level sets in scalar fields and are widely used in topological data analysis and visualization. A main challenge of utilizing contour trees for large-scale scientific data is their computation at scale using high-performance computing. To address this challenge, recent work has introduced distributed hierarchical contour trees for distributed computation and storage of contour trees. However, effective use of these distributed structures in analysis and visualization requires subsequent computation of geometric properties and branch decomposition to support contour extraction and exploration. In this work, we introduce distributed algorithms for augmentation, hypersweeps, and branch decomposition that enable parallel computation of geometric properties, and support the use of distributed contour trees as query structures for scientific exploration. We evaluate the parallel performance of these algorithms and apply them to identify and extract important contours for scientific visualization.

Computational Geometry,Distributed, Parallel, and Cluster Computing

Watershed-based Weighted Graph Partitioning Active Contours

Decompose Image into Meaningful Regions Based on Contour Detector and Watershed Algorithm

Mesh Decomposition With Cross-Boundary Brushes

An experimental comparison of superpixels detection methods for contour detection

IMAGE SEGMENTATION: A WATERSHED TRANSFORMATION ALGORITHM

Adaptive Decomposition and Shared Weight Volumetric Transformer Blocks for Efficient Patch-Free 3D Medical Image Segmentation.

Image Segmentation Based on Watershed Transform

Efficiently Solving the Piecewise Constant Mumford-Shah Model using Graph Cuts

Neighborhood Aided Implicit Active Contours

Watersheds Algorithm for Relative Potential Energy

Voronoi-based Topological Contour Filling Approach for Environmental Impact Assessment

An Improved Contour Detection Method Based on Level Set and Watershed Transform

Image segmentation method based on improved graph cut algorithm

Joint Dense-Point Representation for Contour-Aware Graph Segmentation

Gushing And Immersion Alternative Watershed Algorithm

Watershed-based Textural Image Segmentation

A Fast Watersheds Approach For Image Segmentation Based On Relative Potential Energy

Contour-enhanced Resampling of 3D Point Clouds Via Graphs

Weighted Laplacian Method and Its Theoretical Applications

A Fuzzy Geometric Active Contour Method for Image Segmentation.

Distributed Augmentation, Hypersweeps, and Branch Decomposition of Contour Trees for Scientific Exploration