Contour Position Research Articles

Fuzzy Active Contour Model Using Fractional-Order Diffusion Based Edge Indicator and Fuzzy Local Fitted Image

Local region-based active contour models (ACMs) can effectively segment images corrupted by intensity inhomogeneity, however, they always converge to local minimum and are sensitive to the initial position of contour. In this paper, a novel fuzzy ACM is proposed to tackle these problems. In order to deal with intensity inhomogeneity, the fuzzy local fitted image is first defined and utilized for constructing a local-region based fuzzy energy term, which is minimized in a variational level set framework to accurately segment inhomogeneous images. Second, the fractional-order diffusion based edge indicator is used to scale the local fuzzy energy term to reduce the effect of intensity inhomogeneity. Third, the fuzzy signed pressure force (FSPF) function defined by local image information is used for constructing the weighted area term to further improve the accuracy of the developed model. Finally, the global FSPF is formulated and used as an adaptive force, which can drive the level set function (LSF) to adaptively move up or down according to image intensity information. Therefore, the initial contour can be initialized as a constant function, which eliminates the problem caused by contour initialization. Moreover, the global FSPF makes the proposed model not easy to fall into local minimum. The results of experiments on synthetic and real images validate the accuracy of the proposed model for inhomogeneous image segmentation.

Hybrid SPF and KD Operator-Based Active Contour Model for Image Segmentation

Image segmentation is a crucial stage of image analysis systems because it detects and extracts regions of interest for further processing, such as image recognition and the image description. However, segmenting images is not always easy because segmentation accuracy depends significantly on image characteristics, such as color, texture, and intensity. Image inhomogeneity profoundly degrades the segmentation performance of segmentation models. This article contributes to image segmentation literature by presenting a hybrid Active Contour Model (ACM) based on a Signed Pressure Force (SPF) function parameterized with a Kernel Difference (KD) operator. An SPF function includes information from both the local and global regions, making the proposed model independent of the initial contour position. The proposed model uses an optimal KD operator parameterized with weight coefficients to capture weak and blurred boundaries of inhomogeneous objects in images. Combined global and local image statistics were computed and added to the proposed energy function to increase the proposed model's sensitivity. The segmentation time complexity of the proposed model was calculated and compared with previous state-of-the-art active contour methods. The results demonstrated the significant superiority of the proposed model over other methods. Furthermore, a quantitative analysis was performed using the mini-MIAS database. Despite the presence of complex inhomogeneity, the proposed model demonstrated the highest segmentation accuracy when compared to other methods.

Hybrid Active Contour Based on Local and Global Statistics Parameterized by Weight Coefficients for Inhomogeneous Image Segmentation

Image inhomogeneity often occurs in real-world images and may present considerable difficulties during image segmentation. Therefore, this paper presents a new approach for the segmentation of inhomogeneous images. The proposed hybrid active contour model is formulated by combining the statistical information of both the local and global region-based energy fitting models. The inclusion of the local region-based energy fitting model assists in extracting the inhomogeneous intensity regions, whereas the curve evolution over the homogeneous regions is accelerated by including the global region-based model in the proposed method. Both the local and global region-based energy functions in the proposed model drag contours toward the accurate object boundaries with precision. Each of the local and global region-based parts are parameterized with weight coefficients, based on image complexity, to modulate two parts. The proposed hybrid model is strongly capable of detecting region of interests (ROIs) in the presence of complex object boundaries and noise, as its local region-based part comprises bias field. Moreover, the proposed method includes a new bias field (NBF) initialization and eliminates the dependence over the initial contour position. Experimental results on synthetic and real-world images, produced by the proposed model, and comparative analysis with previous state-of-the-art methods confirm its superior performance in terms of both time efficiency and segmentation accuracy.

Inhomogeneous Image Segmentation Using Hybrid Active Contours Model With Application to Breast Tumor Detection

The most fatal and frequent cancer amongst women is breast cancer. Mammography provides timely detection of lumps and masses in breast tissue, but effective diagnosis requires accurately identifying malignant tumor boundaries, which remains challenging, particularly for images with inhomogeneous regions. Therefore, we propose an active contour method based on a reformed combined local and global fitted function to address breast tumor segmentation. This combined function is strengthened by a proposed average energy driving function to capture obscure boundaries for regions of interest more precisely from inhomogeneous images. Including a p-Laplace term eliminates reinitialization requirements and suppresses false contours in the segmentation. Bias field signal, which causes image homogeneity corruption, is estimated by bias field initialization to ensure independence from the initial contour position. Local and global fitted models are incorporated by introducing bias fields within them. The proposed method was tested on the MIAS MiniMammographic Database, with quantitative analysis to calculate its accuracy, effectiveness, and efficiency. Experimentation confirmed the proposed method provided superior results compared with previous state-of-the-art methods.

Saliency-Driven Active Contour Model for Image Segmentation

Active contour models have achieved prominent success in the area of image segmentation, allowing complex objects to be segmented from the background for further analysis. Existing models can be divided into region-based active contour models and edge-based active contour models. However, both models use direct image data to achieve segmentation and face many challenging problems in terms of the initial contour position, noise sensitivity, local minima and inefficiency owing to the in-homogeneity of image intensities. The saliency map of an image changes the image representation, making it more visual and meaningful. In this study, we propose a novel model that uses the advantages of a saliency map with local image information (LIF) and overcomes the drawbacks of previous models. The proposed model is driven by a saliency map of an image and the local image information to enhance the progress of the active contour models. In this model, the saliency map of an image is first computed to find the saliency driven local fitting energy. Then, the saliency-driven local fitting energy is combined with the LIF model, resulting in a final novel energy functional. This final energy functional is formulated through a level set formulation, and regulation terms are added to evolve the contour more precisely across the object boundaries. The quality of the proposed method was verified on different synthetic images, real images and publicly available datasets, including medical images. The image segmentation results, and quantitative comparisons confirmed the contour initialization independence, noise insensitivity, and superior segmentation accuracy of the proposed model in comparison to the other segmentation models.

Active contour model based on LIF model and optimal DoG operator energy for image segmentation

In order to solve the problem that the region-based active contour model is sensitive to the initial contour position, the convergence is poor, and the active contour model can not obtain good segmentation results when segmenting complex background images and severe intensity inhomogeneous images. In this paper, an active contour model which combines Local Image Fitting (LIF) and Difference of Gaussian (DoG) operator energy for image segmentation is proposed. Firstly, an optimal DoG operator is obtained by using the edge energy term, it can enhance the edge while smoothing inhomogeneous regions. Then, using the DoG energy term which is obtained in the first step and the LIF energy term to construct the total function energy terms. In this process, the regularization term is also established, it can control the smoothness of evolution curve, avoiding over-segmentation and re-initialization step. Finally, the variational method and gradient descent flow method are adopted to minimize the total energy functional for segmentation. Compared with other region-based active contour models, the experimental results show that the proposed method can achieve a better segmentation performance with less iterations and high calculation efficiency while segmenting the synthetic and real images with intensity inhomogeneity.

SEGMENTASI CITRA THORAX PARU-PARU MANUSIA DARI SINAR-X MENGGUNAKAN METODE KONTUR AKTIF

Research on medical image objects in the form of lung images of thoracic X-Rayis increasingly being developed because the information contained in medical images is used to analyze and determine the shape of the lungs. The process of normalization and image improvement is needed and continued with the segmentation process using the right method. The active snake contour method is used because it is resistant to the noise around the object. The research has been usedthe Matlab software GUI program version R2015a. The image through the initial preprocessing stage is converted into a grayscale image. The segmentation process used after the initialization process in the form of a small circle curve placed of the object to be segmented and the determination position of the active contour or detemination of the active parameters of the contour. Determination of the value active contour parameters greatly influences the results of segmentation and influences the direction of active contour movement. If the active coordinate position of the contour is outside the area to be segmented it will cause active contours to move away from the object. Validation the level of accuracy of segmentation results is done by comparing the results of the snake active contour segmentation to the results of manual segmentationused MSE method

Entropy-based global and local weight adaptive image segmentation models

This paper proposes a parameter adaptive hybrid model for image segmentation. The hybrid model combines the global and local information in an image, and provides an automated solution for adjusting the selection of the two weight parameters. Firstly, it combines an improved local model with the global Chan-Vese (CV) model , while the image's local entropy is used to establish the index for measuring the image's gray-level information. Parameter adjustment is then performed by the real-time acquisition of the ratio of the different functional energy in a self-adapting model responsive to gray-scale distribution in the image segmentation process. Compared with the traditional linear adjustment model, which is based on trial-and-error, this paper presents a more quantitative and intelligent method for achieving the dynamic nonlinear adjustment of global and local terms. Experiments show that the proposed model achieves fast and accurate segmentation for different types of noisy and non-uniform grayscale images and noise images. Moreover, the method demonstrates high stability and is insensitive to the position of the initial contour.

Experimental study of the onset of downstream motion of adhering droplets in turbulent shear flows

In this study, the dynamics of adhering liquid droplets on various solid surfaces in shear flow, which is driven by a controlled air flow, are investigated experimentally. A series of experiments are carried out to understand the effects of fluid properties, wetting characteristics, droplet sizes and flow velocities. A rectangular Plexiglas-channel is used for the experiments. Droplets are placed on the bottom wall of that channel. The droplet shape and contour position are measured by the transmitted light technique and are further analyzed by using image processing. The shear flow leads to a deformation and oscillation of the droplet and eventually to a movement downstream. Three regimes are identified: regime I corresponds to a deformation and oscillation, whereas in regime II the deformed and still oscillating droplets moves downstream. Regime III characterizes a continuous downstream movement in a gliding manner. The start of regime II, i.e. the onset of downstream motion is defined by a critical air velocity. Results show that the contact angle hysteresis and surface energy have a strong influence on the critical air velocity for the onset of droplet motion, as well as the wetting characteristics and droplet volume. To find a global empirical law for the critical velocity, a dimensionless approach is derived based on the droplet Reynolds number and a modified Laplace number. The empirical law is in good agreement with experimental data and may serve as an estimator of the critical velocity for the onset of droplet motion.

Level set model for water region segmentation in synthetic aperture radar images

A level set model is presented for water region segmentation in synthetic aperture radar (SAR) images. We formulate the segmentation problem within a global energy minimization framework. First, the background and foreground regions in SAR images are modeled as G0 distributions. They are then used to construct the energy functional for the desired regions. To avoid the local minimum problem, the energy functional is transferred into a strictly convex model that guarantees the existence of the global minimum. During the iterative process, a sinusoidal signed pressure force (SPF) function is applied to efficiently locate weak or blurred edges in the heterogeneous regions. Finally, a Gaussian convolution is used to equivalently substitute the Laplacian of the level set function in the evolution equation, which omits the reinitialization at each iteration. Since based on the stationary global minimum, the presented model can accurately detect inside edges, regardless of the position and shape of the initial contour. Furthermore, because the SPF function can enhance the acquisition ability to the target contour, the internal and external motions of the curve can be accelerated. Thus, the convergence speed of the curve can be improved significantly. The experimental results based on the simulated and real SAR data demonstrate the effectiveness of our method.

A harmonic-cancellation-based model to predict speech intelligibility against a harmonic masker

This study aimed at predicting speech intelligibility in the presence of harmonic maskers. Contrary to a noise signal, these maskers have a harmonic structure that allows for a segregation of the competing sounds based on a difference of their fundamental frequency (F0). This F0 segregation could be due to spectral glimpsing or harmonic cancellation. It is unclear what the relative contributions of these two mechanisms are. In this work, we have modified the model of Collin and Lavandier [J. Acoust. Soc. Am. 134, 1146–1159 (2013)] in order to take into account both mechanisms. Different implementations of the model were compared and applied to two data sets: SRTs measured for monotonous harmonic maskers that varied in their F0s and degree of harmonicity; SRTs measured for monotonous and intonated harmonic maskers that varied in their mean F0, F0 contour, temporal envelope and spatial position. Comparison of data and model predictions will allow us to establish how to implement harmonic cancellation in our model framework, to evaluate to what extent the model can predict F0 segregation for harmonic maskers, and to determine the relative roles of spectral glimpsing and harmonic cancellation suggested by the model.

Dynamic Regulation of Level Set Parameters Using 3D Convolutional Neural Network for Liver Tumor Segmentation.

Segmentation of liver tumors plays an important role in the choice of therapeutic strategies for liver disease and treatment monitoring. In this paper, we generalize the process of a level set with a novel algorithm of dynamic regulation to energy functional parameters. The presented method is fully automatic once the tumor has been detected. First, a 3D convolutional neural network with dense layers for classification is used to estimate current contour location relative to the tumor boundary. Second, the output 3D CNN probabilities can dynamically regulate parameters of the level set functional over the process of segmentation. Finally, for full automation, appropriate initializations and local window size are generated based on the current contour position probabilities. We demonstrate the proposed method on the dataset of MICCAI 2017 LiTS Challenge and 3DIRCADb that include low contrast and heterogeneous tumors as well as noisy images. To illustrate the strength of our method, we evaluated it against the state-of-the-art methods. Compared with the level set framework with fixed parameters, our method performed better significantly with an average DICE improvement of 0.15. We also analyzed a challenging dataset 3DIRCADb of tumors and obtained a competitive DICE of 0.85 ± 0.06 with the proposed method.

Active contour model based on variable exponent p-Laplace equation for image segmentation

ABSTRACTIn this paper, we propose a region-based active contour model for image segmentation. By combining the region fitting energy based on coefficient of variation with the variable exponent p-Laplace energy, the proposed method can perform well in segmenting complex images. The region fitting energy conducts the evolving curve to reach the boundaries of the objects, and the p-Laplace energy can handle the topological changes and extract the boundaries accurately. In order to eliminate the re-initialization step, an augmented Lagrangian method is employed to solve the optimization problem. The results of experiments on synthetic and real images demonstrate that our method can successfully segment complex object boundaries, and it is robust to noise and not sensitive to the initial position of contours.

The motion-induced contour revisited: Observations on 3-D structure and illusory contour formation in moving stimuli.

The motion-induced contour (MIC) was first described by Victor Klymenko and Naomi Weisstein in a series of papers in the 1980s. The effect is created by rotating the outline of a tilted cube in depth. When one of the vertical edges is removed, an illusory contour can be seen in its place. In four experiments, we explored which stimulus features influence perceived illusory contour strength. Participants provided subjective ratings of illusory contour strength as a function of orientation of the stimulus, separation between inducing edges, and the length of inducing edges. We found that the angle of tilt of the object in depth had the largest impact on perceived illusory contour strength with tilt angles of 20° and 30° producing the strongest percepts. Tilt angle is an unexplored feature of structure-from-motion displays. In addition, we found that once the depth structure of the object was extracted, other features of the display, such as the distance spanned by the illusory contour, could also influence its strength, similar to the notion of support ratio for 2-D illusory contours. Illusory contour strength was better predicted by the length of the contour in 3-D rather than in 2-D, suggesting that MICs are constructed by a 3-D process that takes as input initially recovered contour orientation and position information in depth and only then forms interpolations between them.

Automatic Building Extraction from Image and LiDAR Data with Active Contour Segmentation

Automatic building extraction is an important topic for many applications such as urban planning, disaster management, 3D building modeling and updating GIS databases. Its approaches mainly depend on two data sources: light detection and ranging (LiDAR) point cloud and aerial imagery both of which have advantages and disadvantages of their own. In this study, in order to benefit from the advantages of each data sources, LiDAR and image data combined together. And then, the building boundaries were extracted with the automated active contour algorithm implemented in MATLAB. Active contour algorithm uses initial contour positions to segment an object in the image. Initial contour positions were detected without user interaction by a series of image enhancements, band ratio and morphological operations. Four test areas with varying building and background levels of detail were selected from ISPRS’s benchmark Vaihingen and Istanbul datasets. Vegetation and shadows were removed from all the datasets by band ratio to improve segmentation quality. Subsequently, LiDAR point cloud data was converted to raster format and added to the aerial imagery as an extra band. Resulting merged image and initial contour positions were given to the active contour algorithm to extract building boundaries. In order to compare the contribution of LiDAR to the proposed method, the boundaries of the buildings were extracted from the input image before and after adding LiDAR data to the image as a layer. Finally extracted building boundaries were smoothed by the Awrangjeb (Int J Remote Sen 37(3): 551–579. https://doi.org/10.1080/01431161.2015.1131868 , 2016) boundary regularization algorithm. Correctness (Corr), completeness (Comp) and accuracy (Q) metrics were used to assess accuracy of segmented building boundaries by comparing extracted building boundaries with manually digitized building boundaries. Proposed approach shows the promising results with over 93% correctness, 92% completeness and 89% quality.

Dynamic Block Matching to assess the longitudinal component of the dense motion field of the carotid artery wall in B-mode ultrasound sequences - Association with coronary artery disease.

The motion of the common carotid artery tissue layers along the vessel axis during the cardiac cycle, observed in ultrasound imaging, is associated with the presence of established cardiovascular risk factors. However, the vast majority of the (semi-)automatic methods devised to measure this so-called "longitudinal kinetics" phenomenon are based on the tracking of a single point, thus failing to capture the overall - and potentially inhomogeneous - motion of the entire arterial wall. The aim of this work is to introduce a motion tracking a framework to simultaneously extract the temporal trajectory of a large collection of points (several hundred) horizontally aligned and spanning the entire exploitable width of the image, thus providing a dense motion field. The only action required from the user is to indicate the left and right borders of the region to be processed. A previously validated contour segmentation method is used to position one point in the arterial wall in each column of the image. Between two consecutive frames, the radial motion of each point is predetermined by the position of the segmentation contours. The longitudinal motion, which is the main focus of the present work, is determined in two steps. First, a series of independent block matching operations are carried out for all the tracked points. Here, the displacement of each point is not determined yet, instead the similarity map is stored. Then, an original dynamic-programming approach is exploited to regularize the collection of similarity maps and estimate the globally optimal motion over the entire vessel wall. Sixty-two atherosclerotic participants at high cardiovascular risk were involved in this study. Method training and validation was performed with 20 and 42 participants, respectively. The amplitude-independent index σX was introduced to quantitate the motion inhomogeneity across the length of the artery. A dense displacement field, describing the longitudinal motion of the carotid far wall over time, was extracted from all participants. For each cine-loop, the method was evaluated against manual reference tracings performed on three local points, and showed a good accuracy, with an average absolute error (± STD) of 150 (±163)μm. It also demonstrated an overall greater robustness compared to a previously validated method based on single-point motion tracking. For all the 62 participants, the analyzed region had, in average, a width of 24.2mm, involving the simultaneous tracking of 357 points along 151 temporal frames, and requiring a total computational time of 68 s. Analyzing the inhomogeneity of the carotid artery motion showed a strong correlation between σX and the presence of coronary artery disease (β-coefficient=0.586, P=0.003). To the best of our knowledge, this is the first time that a method is specifically proposed to assess the dense motion field corresponding to the longitudinal kinetics of the carotid far wall. This approach has potential to evaluate the homogeneity (or lack thereof) of the wall dynamics. The proposed method has promising performances to improve the analysis of arterial longitudinal motion and the understanding of the underlying patho-physiological parameters.

Intrafraction motion quantification and planning target volume margin determination of head-and-neck tumors using cine magnetic resonance imaging

PurposeTo quantify intrafractional motion to determine population-based radiotherapy treatment margins for head-and-neck tumors. MethodsCine MR imaging was performed in 100 patients with head-and-neck cancer on a 3T scanner in a radiotherapy treatment setup. MR images were analyzed using deformable image registration (optical flow algorithm) and changes in tumor contour position were used to calculate the tumor motion. The tumor motion was used together with patient setup errors (450 patients) to calculate population-based PTV margins. ResultsTumor motion was quantified in 84 patients (12/43/29 nasopharynx/oropharynx/larynx, 16 excluded). The mean maximum (95th percentile) tumor motion (swallowing excluded) was: 2.3 mm in superior, 2.4 mm in inferior, 1.8 mm in anterior and 1.7 mm in posterior direction. PTV margins were: 2.8 mm isotropic for nasopharyngeal tumors, 3.2 mm isotropic for oropharyngeal tumors and 4.3 mm in inferior–superior and 3.2 mm in anterior–posterior for laryngeal tumors, for our institution. ConclusionsIntrafractional head-and-neck tumor motion was quantified and population-based PTV margins were calculated. Although the average tumor motion was small (95th percentile motion <3.0 mm), tumor motion varied considerably between patients (0.1–12.0 mm). The intrafraction motion expanded the CTV-to-PTV with 1.7 mm for laryngeal tumors, 0.6 mm for oropharyngeal tumors and 0.2 mm for nasopharyngeal tumors.

A mathematical model for the 3d location estimation of 2D echocardiography data

The recovery of 3D left ventricle(LV) shape using 2D echocardiography is very attractable topic in the field of ultrasound imaging. In this paper, we propose a mathematical model to determine the 3D position of LV contours extracted from multiple 2D echocardiography images. We formulate the proposed model as a non-convex constrained minimization problem. To solve it, we propose a proximal alternating minimization algorithm with a solver OPTI for quadratically constrained quadratic program. For validating the proposed model, numerical experiments are performed with real ultrasound data. The experimental results show that the proposed model is promising and available for real echocardiography data.

A Bayesian Approach to Predict Sub-Annual Beach Change and Recovery

The upper beach, between the astronomical high tide and the dune-toe, supports habitat and recreation along many beaches, making predictions of upper beach change valuable to coastal managers and the public. We developed and tested a Bayesian network (BN) to predict the cross-shore position of an upper beach elevation contour (ZlD) following 1 month to 1-year intervals at Fire Island, New York. We combine hydrodynamic data with series of island-wide topographic data and spatially limited cross-shore profiles. First, we predicted beach configuration of ZlD positions at high spatial resolution (50 m) over intervals spanning 2005–2014. Compared to untrained model predictions, in which all six outcomes are equally likely (prior likelihood = 0.16), our prediction metrics (skill = 0.52; log likelihood ratio = 0.14; accuracy = 0.56) indicate the BN confidently predicts upper beach dynamics. Next, the BN forecasted three intervals of beach recovery following Hurricane Sandy. Results suggest the pre-Sandy training data is sufficiently robust to require only periodic updates to beach slope observations to maintain confidence for forecasts. Finally, we varied input data, using observations collected at a range of temporal (1–12 months) and spatial (50 m to > 1 km) resolutions to evaluate model skill. This experiment shows that data collection techniques with different spatial and temporal frequencies can be used to inform a single modeling framework and can provide insight to BN training requirements. Overall, results indicate that BNs and inputs can be developed for broad coastal change assessment or tailored to a set of predictive requirements, making this methodology applicable to a variety of coastal prediction scenarios.

An Efficient Multi-Scale Local Binary Fitting-Based Level Set Method for Inhomogeneous Image Segmentation

An efficient level set model based on multiscale local binary fitting (MLBF) is proposed for image segmentation. By introducing multiscale idea into the LBF model, the proposed MLBF model can effectively and efficiently segment images with intensity inhomogeneity. In addition, by adding a reaction diffusion term into the level set evolution (LSE) equation, the regularization of the level set function (LSF) can be achieved, thus completely eliminating the time-consuming reinitialization process. In the implementation phase, in order to greatly improve the efficiency of the numerical solution of the level set segmentation model, we introduce three strategies: The first is the additive operator splitting (AOS) solver which is used for breaking the restrictions on time step; the second is the salient target detection mechanism which is used to achieve full automatic initialization of the LSE process; the third is the sparse filed method (SFM) which is used to restrict the groups of pixels that need to be updated in a small strip region. Under the combined effect of these three strategies, the proposed model achieves very high execution efficiency in the following aspects: contour location accuracy, speed of evolution convergence, robustness against initial contour position, and robustness against noise interference.