Region-based Level Set Method Research Articles

Multi-scale local region based level set method for image segmentation in the presence of intensity inhomogeneity

Intensity inhomogeneity arising from the imperfect image acquisition process is a major challenge for image segmentation. Most of widely used image segmentation methods usually fail to segment the image with intensity inhomogeneity due to the assumption of intensity homogeneity. In this paper, an efficient multi-scale local region based level set method is proposed to segment the image with intensity inhomogeneity, which is based on the multi-scale segmentation and statistical analysis for intensities of local region. Firstly, the local region is defined in circular shape for capturing more local intensity information. The statistical analysis can be performed on intensities of local circular regions centered in each pixel by using multi-scale low-pass filtering. Then, the data term of level set energy functional can be constructed by approximating the normalized weighted image divided by multi-scale local intensity information in a piecewise constant way. In addition, the regularization term is built to control the smoothness of evolving curve and avoid the over-segmentation phenomenon and re-initialization step. Finally, the multi-scale segmentation is performed by minimizing the total level set energy functional by using the finite difference scheme. The experiments on synthetic and real images with slight or severe intensity inhomogeneity can demonstrate the efficiency and robustness of the proposed method. In addition, the comparisons with the recently popular local binary fitting (LBF) model and local Chan-Vese (LCV) model also show that our method has obvious superiority over the traditional local region based methods.

A Hybrid Method Based on Fuzzy Clustering and Local Region-Based Level Set for Segmentation of Inhomogeneous Medical Images

medical images are more affected by intensity inhomogeneity rather than noise and outliers. This has a great impact on the efficiency of region-based image segmentation methods, because they rely on homogeneity of intensities in the regions of interest. Meanwhile, initialization and configuration of controlling parameters affect the performance of level set segmentation. To address these problems, this paper proposes a new hybrid method that integrates a local region-based level set method with a variation of fuzzy clustering. Specifically it takes an information fusion approach based on a coarse-to-fine framework that seamlessly fuses local spatial information and gray level information with the information of the local region-based level set method. Also, the controlling parameters of level set are directly computed from fuzzy clustering result. This approach has valuable benefits such as automation, no need to prior knowledge about the region of interest (ROI), robustness on intensity inhomogeneity, automatic adjustment of controlling parameters, insensitivity to initialization, and satisfactory accuracy. So, the contribution of this paper is to provide these advantages together which have not been proposed yet for inhomogeneous medical images. Proposed method was tested on several medical images from different modalities for performance evaluation. Experimental results approve its effectiveness in segmenting medical images in comparison with similar methods.

A novel region-based level set method initialized with mean shift clustering for automated medical image segmentation

Appropriate initialization and stable evolution are desirable criteria to satisfy in level set methods. In this study, a novel region-based level set method utilizing both global and local image information complementarily is proposed. The global image information is extracted from mean shift clustering without any prior knowledge. Appropriate initial contours are obtained by regulating the clustering results. The local image information, as extracted by a data fitting energy, is employed to maintain a stable evolution of the zero level set curves. The advantages of the proposed method are as follows. First, the controlling parameters of the evolution can be easily estimated by the clustering results. Second, the automaticity of the model increases because of a reduction in computational cost and manual intervention. Experimental results confirm the efficiency and accuracy of the proposed method for medical image segmentation.

An improved level set method for vertebra CT image segmentation

BackgroundClinical diagnosis and therapy for the lumbar disc herniation requires accurate vertebra segmentation. The complex anatomical structure and the degenerative deformations of the vertebrae makes its segmentation challenging.MethodsAn improved level set method, namely edge- and region-based level set method (ERBLS), is proposed for vertebra CT images segmentation. By considering the gradient information and local region characteristics of images, the proposed model can efficiently segment images with intensity inhomogeneity and blurry or discontinuous boundaries. To reduce the dependency on manual initialization in many active contour models and for an automatic segmentation, a simple initialization method for the level set function is built, which utilizes the Otsu threshold. In addition, the need of the costly re-initialization procedure is completely eliminated.ResultsExperimental results on both synthetic and real images demonstrated that the proposed ERBLS model is very robust and efficient. Compared with the well-known local binary fitting (LBF) model, our method is much more computationally efficient and much less sensitive to the initial contour. The proposed method has also applied to 56 patient data sets and produced very promising results.ConclusionsAn improved level set method suitable for vertebra CT images segmentation is proposed. It has the flexibility of segmenting the vertebra CT images with blurry or discontinuous edges, internal inhomogeneity and no need of re-initialization.

Multi-agent stochastic level set method in image segmentation

A stochastic structure for single and multi-agent level set method is investigated in this article in an attempt to overcome local optima problems in image segmentation. Like other global optimization methods that take advantage of random operators and multi-individual search algorithms, the best agent in this proposed algorithm plays the role of leader in order to enable the algorithm to find the global solution. To accomplish this, the procedure employs a set of stochastic partial differential equations (SPDE), each one of which evolves based on its own stochastic dynamics. The agents are then compelled to simultaneously converge to the best available topology. Moreover, the stochastic dynamics of each agent extends the stochastic level set approach by using a multi source structure. Each source is a delta function centered on a point of evolving front. Lastly, while the computational costs of these methods are higher than the region-based level set method, the probability of finding the global solution is significantly increased.

Region-based image segmentation with local signed difference energy

Intensity inhomogeneity often causes considerable difficulties in image segmentation. To tackle this problem, we propose a new region-based level set method. The proposed method considers the local image information by describing it as a novel local signed difference (LSD) energy, which possesses both local separability and global consistency. The LSD energy term is integrated into an objective energy functional, which is minimized via a level set evolution process. Extensive experiments are performed to evaluate the proposed method, showing improvements in both accuracy and efficiency, as compared with the state-of-the-art approaches.

An Edge-and Region-Based Level Set Method with Shape Priors for Image Segmentation

Traditional level set methods could not deal with the segmentation of overlapped objects and the ones having the similar gray value with background,since these methods just take image data into consideration.Aiming for this problem,we propose a new edge-and region-based level set method for image segmentation.Our method firstly fuses the regional information into the edge-based level set method,and then incorporates shape priors to design a new level set method with shape priors.The comparison results show that the proposed method can segment the overlapped objects better and also achieve better performance on the objects with the similar gray value as background since it integrates multiple kinds of information together.

Three-dimensional surface reconstruction of tree canopy from lidar point clouds using a region-based level set method

In this article, a novel method is proposed for three-dimensional (3D) canopy surface reconstruction of trees using a region-based level set method. Both individual tree crowns and clusters of trees are first marked for further exploration. Multiple horizontal slices corresponding to different heights are obtained. The 3D structure of tree canopy is built using raw data from lidar point clouds. Also, new applications are proposed based on the new method for 3D forest reconstruction. The biomass parameters of the forest, including tree intersection area, tree equivalent crown radius, and canopy volume, can be calculated from stacking 2D slices of trees. Tree types are also identified and classified. The results indicate that this approach is effective for 3D surface reconstruction of forests including individual trees and clusters of trees, and that critical forest parameters (such as tree intersection area, tree position, and canopy volume) can be derived for the evaluation and measurement of biophysical parameters of forests.

Active contours with selective local or global segmentation property for multiobject image

In this paper, a novel region-based active contour model is proposed. Unlike the Chan-Vese model, the energy function of our model only uses information outside the evolving curve. A restriction matrix is introduced for restricting the curve to evolve in the desired region. By minimizing the proposed energy function, the corresponding level set function is obtained. The proposed method has four advantages over the traditional region-based level set methods. First, the level set function is no longer required to be initialized as a sign distance function. Second, due to the simplified level set formulation and its fast curve evolution speed, the level set function does not need to be re-initialized and the computational cost is low. Third, our model is applicable to multiobject images since it can segment objects by separating background from an image. Finally, the proposed model has the property of selective local or global segmentation; it can segment not only the desired object but also the other objects in an image. The advantages of our model are demonstrated using synthetic and real images. C 2011 Society of Photo-Optical

Segmentation of Cereal Plant Images Using Level Set Methods – A Comparative Study

In this paper we evaluate the quality of segmentation of plant images achieved by different level set methods commonly used in the literature. The plants of study are narrow-leaf cereal plants at different growth stages and the segmentation quality measure was considered to be the boundary, leaf tips, and axils. The results show that region-based level set methods can perform the segmentation of plants with high accuracy when the plants are at either early or mature stages of growth. The results also show that contour based level set algorithms are not applicable to the segmentation of narrow leaf plants because the front being computed does not advance to the high curvature features, such as sharp tips and axils. A typical image of a mature plant has isolated regions from the interlacing of leaves. Only region-based methods can perform the segmentation with good accuracy. Level set methods are sensitive to initialization and parameter selection.

Automatic Image Segmentation Method Using Sequential Level Set

为了实现图像的完全分割,基于无须重新初始化的水平集方法提出了一种接力水平集方法.该方法在待分割图像中自动交替地创建嵌套子区域和相应的初始水平集函数,使水平集函数在其中演化并收敛,然后重复这个过程直到子区域面积为0.与原始算法及经典的基于区域的水平集方法相比,该方法具有如下优点:1) 自动完成,无须交互式的初始化;2) 多次分割图像,能够比原始算法检测到更多的边缘;3) 对于非匀质的图像,能够取得比经典的基于区域的水平集方法更好的分割效果;4) 提供一个开放的分割算法框架,其他单水平集方法稍作修改后也可替换这里所使用的单水平集方法.实验结果表明,此算法对人造图像和医学影像实现了无须交互的完全分割,对非匀质图像分割表现出更好的鲁棒性.;Based on the level set method without re-initialization, a sequential level set method is proposed to realize full image segmentation. The proposed method automatically and alternatively creates nested sub-regions or the corresponding initial level set functions in the image to be segmented, and then makes the level set function evolved to be convergence in the corresponding sub-region. This step is sequentially repeated until the sub-region vanishes. Compared with the original method and a representative region-based level set method, the proposed method has many advantages as follows: 1) It is automatically executed and does not need the interactive initialization anymore; 2) It segments image more than once and detects more boundaries than the original method; 3) It can get better performance on non-homogenous images than the representative region-based level set method; 4) It is an open image segmentation framework in which the single level set method is used can be replaced by other single level set methods after some modification. Experimental results indicate that the proposed method could fully segment the synthetic and medical images without interactive step and additionally works more robust on non-homogenous images.

Parallel Computation of the Region-Based Level Set Method for Boundary Detection of Moving Objects

We formulate a parallel, region-based level set model to speed up accurate boundary detection of moving objects in low-contrast images, applying parallelization and discretization to a Chan-Vese (CV) model. We implement the model in a column parallel vision (CPV) system that is one of parallel image processing systems we developed for robot vision. Using a microscopic image of moving paramecia as a sample of a low-contrast image, our model detects moving paramecia boundaries within 2 ms per image. Comparisons of our model to a CV model using the CPV system and a nonparallel PC, we found that our model cuts calculation time for a CV model while obtaining accuracy similar to the CV model in boundary detection of moving objects.

A Unified Tensor Level Set for Image Segmentation

This paper presents a new region-based unified tensor level set model for image segmentation. This model introduces a three-order tensor to comprehensively depict features of pixels, e.g., gray value and the local geometrical features, such as orientation and gradient, and then, by defining a weighted distance, we generalized the representative region-based level set method from scalar to tensor. The proposed model has four main advantages compared with the traditional representative method as follows. First, involving the Gaussian filter bank, the model is robust against noise, particularly the salt- and pepper-type noise. Second, considering the local geometrical features, e.g., orientation and gradient, the model pays more attention to boundaries and makes the evolving curve stop more easily at the boundary location. Third, due to the unified tensor pixel representation representing the pixels, the model segments images more accurately and naturally. Fourth, based on a weighted distance definition, the model possesses the capacity to cope with data varying from scalar to vector, then to high-order tensor. We apply the proposed method to synthetic, medical, and natural images, and the result suggests that the proposed method is superior to the available representative region-based level set method.

Variational Segmentation of Image Sequences Using Region-Based Active Contours and Deformable Shape Priors

In this paper we address the problem of segmentation in image sequences using region-based active contours and level set methods. We propose a novel method for variational segmentation of image sequences containing nonrigid, moving objects. The method is based on the classical Chan-Vese model augmented with a novel frame-to-frame interaction term, which allow us to update the segmentation result from one image frame to the next using the previous segmentation result as a shape prior. The interaction term is constructed to be pose-invariant and to allow moderate deformations in shape. It is expected to handle the appearance of occlusions which otherwise can make segmentation fail. The performance of the model is illustrated with experiments on synthetic and real image sequences.