Showing papers on "Vector flow published in 2011"
TL;DR: The proposed segmentation technique is robust and applicable on various kinds of noisy images without prior knowledge of noise properties and yields better performance than the classical contour models.
Abstract: Finding the correct boundary in noisy images is still a difficult task. This paper introduces a new edge following technique for boundary detection in noisy images. Utilization of the proposed technique is exhibited via its application to various types of medical images. Our proposed technique can detect the boundaries of objects in noisy images using the information from the intensity gradient via the vector image model and the texture gradient via the edge map. The performance and robustness of the technique have been tested to segment objects in synthetic noisy images and medical images including prostates in ultrasound images, left ventricles in cardiac magnetic resonance (MR) images, aortas in cardiovascular MR images, and knee joints in computerized tomography images. We compare the proposed segmentation technique with the active contour models (ACM), geodesic active contour models, active contours without edges, gradient vector flow snake models, and ACMs based on vector field convolution, by using the skilled doctors' opinions as the ground truths. The results show that our technique performs very well and yields better performance than the classical contour models. The proposed method is robust and applicable on various kinds of noisy images without prior knowledge of noise properties.
95 citations
TL;DR: The presented approach represents a generalization for saddle‐type critical points and their separatrices to unsteady vector fields based on generalized streak lines, with the classical vector field topology as its special case for steady vector fields.
Abstract: This paper presents an approach to a time-dependent variant of the concept of vector field topology for 2-D vector fields. Vector field topology is defined for steady vector fields and aims at discriminating the domain of a vector field into regions of qualitatively different behaviour. The presented approach represents a generalization for saddle-type critical points and their separatrices to unsteady vector fields based on generalized streak lines, with the classical vector field topology as its special case for steady vector fields. The concept is closely related to that of Lagrangian coherent structures obtained as ridges in the finite-time Lyapunov exponent field. The proposed approach is evaluated on both 2-D time-dependent synthetic and vector fields from computational fluid dynamics.
50 citations
18 Sep 2011
TL;DR: The proposed method allows curved segmentation paths and thus it is able to segment overlapping characters and touching characters due to low contrast and complex background and helps to improve binarization results, which lead to a better character recognition rate.
Abstract: In this paper, we propose a method based on gradient vector flow for video character segmentation. By formulating character segmentation as a minimum cost path finding problem, the proposed method allows curved segmentation paths and thus it is able to segment overlapping characters and touching characters due to low contrast and complex background. Gradient vector flow is used in a new way to identify candidate cut pixels. A two-pass path finding algorithm is then applied where the forward direction helps to locate potential cuts and the backward direction serves to remove the false cuts, i.e. those that go through the characters, while retaining the true cuts. Experimental results show that the proposed method outperforms an existing method on multi-oriented English and Chinese video text lines. The proposed method also helps to improve binarization results, which lead to a better character recognition rate.
43 citations
1 Dec 2011
TL;DR: This paper proposes an approach for the segmentation of the vascular system of the liver based on the gradient vector flow (GVF) and Frangis vesselness measure, which avoids multi-scale analysis and related scale space problems.
Abstract: An accurate segmentation of vascular systems is fundamental for many medical applications. Stability against different contrast levels and noise are very important. In this paper we propose an approach for the segmentation of the vascular system of the liver. It is based on the gradient vector flow (GVF) and Frangis vesselness measure. This method avoids multi-scale analysis and related scale space problems. It was evaluated on ten CT data-sets.
5 citations
26 Mar 2011
TL;DR: This paper presents a new method for extracting point cloud of useful main surface from 3D point clouds of an existing product or part using gradient vector flow (GVF).
Abstract: This paper presents a new method for extracting point cloud of useful main surface from 3D point cloud of an existing product or part. The 3D point cloud is sliced into 2D point cloud at different slices. For each slice, gradient vector flow (GVF) is generated. An active contour or snake is applied to capture the point cloud belonging to main surface. Points belonging to mini-structures or mini-features are repaired. Experimental results demonstrate the robustness and the effectiveness of the proposed algorithm.
3 citations
9 Jun 2011
TL;DR: A new active contour-based segmentation algorithm which combines external force field ideas and local region based methods in a consistent way is presented, which not only has a large capture range but also can distinguish small details as local regionbased methods do.
Abstract: Motivated by the analysis of knee MRI data arising in the study of osteoarthritis, this paper presents a new active contour-based segmentation algorithm which combines external force field ideas and local region based methods in a consistent way. The approach not only has a large capture range as is common with curve evolution techniques based on static force fields such as the gradient vector flow (GVF) and vector field convolution (VFC) methods, but also can distinguish small details as local region based methods do. The feasibility of the new algorithm is demonstrated on both synthetic images as well as real knee MRI data where the goal is to identify the tibia and femur as part of a larger osteoarthritis image analysis problem.
3 citations
12 Aug 2011
TL;DR: A new optical flow model is proposed which incorporates the harmonic smoothness constraint borrowed from the harmonic gradient vector flow (HGVF) model into the oriented Smoothness constraint and combines the curl term of the harmonic constraint with the oriented smoothness to control the direction of the displacement vectors together.
Abstract: Computation of the optical flow from a sequence of images remains open in the community of computer vision. Two classical models for this problem are the global smoothness algorithm proposed by Horn-Schunck and the oriented smoothness algorithm by Nagel and Enkelmann. In order to increase the accuracy of motion discontinuity, we propose a new optical flow model which incorporates the harmonic smoothness constraint borrowed from the harmonic gradient vector flow (HGVF) model into the oriented smoothness constraint. In particular, we combine the curl term of the harmonic constraint with the oriented smoothness to control the direction of the displacement vectors together and introduce two spatially varying weighting functions to control the above-mentioned two terms. The benefit of the suggested strategies is illustrated qualitatively on the synthetical image and quantitatively on the Middlebury optical flow benchmark. Compared with the classical Horn-Schunck and Nagel-Enkelmann methods, this method can provide more accurate estimation of optical flow around motion discontinuities.
3 citations
8 Jan 2011
TL;DR: A new approach to adaptive estimate of gradient vector flow (GVF) deformable contour based on B-spline representation is proposed, based on the improved dynamic GVF force field which can increase the external force’s capture range and convergence speed.
Abstract: In this paper we proposed and demonstrated a new approach to adaptive estimate of gradient vector flow (GVF) deformable contour based on B-spline representation. An extension of the GVF deformable model is presented, and the method is based on the improved dynamic GVF force field which can increase the external force’s capture range and convergence speed. Then, a specific strategy of adaptive deformable contour knots insertion process, based on fitting accuracy, are proposed, and this approach can automatically add knots in the contour curve according to the reconstruction error analysis. The improved iterative algorithm can reduce the iterative number, and increase fitting accuracy and efficiency. Finally, using computer simulation, the experiments reported in this paper demonstrate an efficient procedure and fine performance of the approach.
2 citations
12 Aug 2011
TL;DR: Wang et al. as discussed by the authors used gradient vector flow as vector field and introduced a vesselness measure to detect vessel which gives high and homogeneous output for line structure so that it is more suitable for segmentation over Frangi's vesselness measures.
Abstract: Vascular diseases are major public heath problem around the world. Vessel segmentation has been widely concerned because it is a key step for diagnosis and surgical planning. Among past strategies, multi-scale line filters are very popular detectors. However, multi-scale integration results in undesirable diffusion when two vessels are closely located. To avoid this problem, we use gradient vector flow as vector field and introduce a vesselness measure to detect vessel which gives high and homogeneous output for line structure so that it is more suitable for segmentation over Frangi's vesselness measure. Level set method is applied to perform vessel segmentation. Our model is tested on real images. Experimental results demonstrate that our approach can successfully separate closely adjacent vessels and address the problems of low contrast and varying vessel width. It shows better performance than multi-scale approach. Furthermore, gradient vector flow makes the contour moving into boundary concavities.
2 citations
3 Dec 2011
TL;DR: A novel external force called adaptively normal biased gradient vector flow (ANBGVF) for active contours, which adaptively generates the diffusion along the tangential direction of the isophotes and biases that along the normal direction is proposed.
Abstract: Gradient vector flow (GVF) is an effective external force for active contours, but its isotropic nature handicaps its performance. The recently proposed NGVF model is an isotropic since it only keeps the diffusion along the normal direction of the isophotes, however, it is sensitive to noise and could erase weak boundaries. In this paper, we propose a novel external force called adaptively normal biased gradient vector flow (ANBGVF) for active contours, which adaptively generates the diffusion along the tangential direction of the isophotes and biases that along the normal direction. Consequently, the ANBGVF snake can preserve weak edges and smooth out noise while maintaining other desirable properties of GVF and NBGVF, such as enlarged capture range, initialization insensitivity and good convergence at concavities. We demonstrate the advantages on synthetic and real images.
1 citations
1 Oct 2011
TL;DR: This paper presents the team's initial efforts in developing a high frame rate vector flow imaging framework that is based on plane wave excitation principles and a high dynamic range block matching algorithm that incorporates least squares fitting principles.
Abstract: Analysis of the complex blood flow pattern in the carotid bifurcation is clinically important to the diagnosis of carotid stenoses. We hypothesize that the use of high frame rate imaging methods such as plane wave excitation, together with vector flow estimators like block matching, may potentially be a suitable imaging problem to this problem. This paper presents our team's initial efforts in developing a high frame rate vector flow imaging framework that is based on plane wave excitation principles and a high dynamic range block matching algorithm that incorporates least squares fitting principles. We have conducted a series of Field II simulations on straight tubes and carotid bifurcation to evaluate the estimation accuracy and imaging performance of our framework. Results indicate that high-frame-rate vector flow imaging is capable of visualizing complex blood flow. It has potential to be further developed into a new clinical technique for vascular diagnoses.
TL;DR: In this paper, a new pressure-like force that not only improves contour convergence rate, but also encourages contours to conform to concave regions is proposed, which is steerable according to the image content.
Abstract: Active contour models have been widely used in various image analysis applications. Despite their usefulness, there are
problems limiting their utility, such as capture range, concavity conformation, and convergence rate. This paper presents
a new pressure-like force that not only improves contour convergence rate, but also encourages contours to conform to
concave regions. Unlike the traditional pressure force, this new force does not require users' input for the force direction
and is steerable according to the image content. Better convergence rate as well as force normalization consistency of this
new force are presented when compared with those of the gradient vector flow force field on synthetic images. Accuracies
of these two methods are compared against the manual markups on a set of cardiac MRI images. Moreover, results on a
MRI image smoothed at different levels demonstrate the robustness of this new force to noise.
20 Jun 2011
TL;DR: This paper describes procedures employing proven active contours and evolutionary algorithms for recognizing points of interest in the images that may serve in determining various parameters and properties of analyzed objects and compares their approach with common numerical methods on real industrial images segmentation.
Abstract: In this paper we describe the extension of system FOTOM capabilities with respect to segmentation of specific mining images. We focus on methods that are inherently resistant against noise present in experimental pit at VSB Technical University. Here, we describe procedures employing proven active contours and evolutionary algorithms for recognizing points of interest in the images that may serve in determining various parameters and properties of analyzed objects. We use the evolutionary algorithms to optimize the parameters of the gradient vector flow field and the parameters affecting the geometrical properties of closed curve used to approximate the location and shape of object boundaries. We suppose that evolutionary algorithms can be used to find the desired global solution. As the computation of gradient vector flow field and also the evolution of active contour are computationally very expensive, we incorporate the GPU acceleration. In conclusion, we compare our approach with common numerical methods on real industrial images segmentation.
TL;DR: A new mean shift based gradient vector flow (GVF) algorithm that drives the internal/external energies towards the correct direction within the standard GVF cost function is introduced.
TL;DR: A Riemannian variational formalism is proposed, and the proposed methodology is illustrated with synthetic and empirical examples of optical-flow vector field decompositions obtained on a variety of surface objects.
1 Oct 2011
TL;DR: This study presents the first quantification and visualization of secondary flow patterns with vector flow ultrasound, using the first commercial implementation of the vector flow method Transverse Oscillation to obtain in-vivo, 2D vector fields in real-time.
Abstract: This study presents the first quantification and visualization of secondary flow patterns with vector flow ultrasound. The first commercial implementation of the vector flow method Transverse Oscillation was used to obtain in-vivo, 2D vector fields in real-time. The hypothesis of this study was that the rotational direction is constant within each artery. Three data sets of 10 seconds were obtained from three main arteries in healthy volunteers. For each data set the rotational flow patterns were identified during diastole. Each data set contains a 2D vector field over time using the vector angles and velocity magnitudes the blood flow patterns were visualized using streamlines in Matlab (Mathworks, Natick, MA, USA). The rotational flow was quantified by the angular frequency for each cardiac cycle, and the mean rotational frequencies and standard deviations were calculated for the abdominal aorta {-1.3±0.4;-1.0±0.3;-0.9±0.2}Hz, the common iliac artery {-0.4±0.1;-1.0±0.2;-0.4±0.1}Hz, and the common carotid artery {0.8±0.3;1.4±0.3;0.4±0.1}Hz. A positive sign indicates an anti-clockwise rotation, and a negative sign indicates clockwise rotation. The sign of the rotational directions within each artery were constant.