Markov Random Field Method Research Articles

A fast, weighted CRF algorithm based on a two-step superpixel generation for SAR image segmentation

ABSTRACT Among the Markov random field methods, conditional random fields (CRF) model has shown good results in Synthetic Aperture Radar (SAR) image segmentation. CRF not only directly models the posterior distribution of label field conditioned on images but also allows the interactions between observations. In this paper, we propose a Fast, Weighted CRF (FWCRF) algorithm based on a two-step superpixel generation method. In our method, heterogeneity of intensity in SAR images due to speckle noise and backscattering is considered. The first step is a preprocessing task for suppressing speckle noise which is done by the L 0 smoothing algorithm. It is followed by Sobel edge detector to highlight heterogeneous regions and edges. Then SAR image is partitioned into homogeneous and heterogeneous regions by using fast robust fuzzy c-means clustering (FRFCM). Simultaneously, the simple linear iterative clustering (SLIC) algorithm is applied to split the image into superpixels. Next, the superpixels belonging to the same class are merged by using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method. Finally, the SAR image is labelled using the FWCRF algorithm which consists of weighted pairwise potential based on adaptive features and applying FRFCM as an initial segmentation method. The proposed algorithm is evaluated in comparison with other conditional random field schemes. The results of our proposed method demonstrate accuracy improvement in segmentation.

An Improved Non-Parametric Method for Multiple Moving Objects Detection in the Markov Random Field

Detecting moving objects in the stationary background is an important problem in visual surveillance systems. However, the traditional background subtraction method fails when the background is not completely stationary and involves certain dynamic changes. In this paper, according to the basic steps of the background subtraction method, a novel non-parametric moving object detection method is proposed based on an improved ant colony algorithm by using the Markov random field. Concretely, the contributions are as follows: 1) A new nonparametric strategy is utilized to model the background, based on an improved kernel density estimation; this approach uses an adaptive bandwidth, and the fused features combine the colours, gradients and positions. 2) A Markov random field method based on this adaptive background model via the constraint of the spatial context is proposed to extract objects. 3) The posterior function is maximized efficiently by using an improved ant colony system algorithm. Extensive experiments show that the proposed method demonstrates a better performance than many existing state-of-the-art methods.

Robust and Accurate Automated Methods for Detection and Segmentation of Brain Tumor in MRI

In this proposed study, a novel Multimodal brain MR image segmentation method is presented to overcome the unattractive and undesirable over segmentation characteristics of conventional Watershed method. The proposed work, presents Optimal Region Amalgamation Technique (RAT) that merge the Watershed method (spatial domain) and Fuzzy C-means clustering (feature spaces) to reduce the unattractive and undesirable over segmentation in brain MR images. In the proposed work, to improve the quality of segmentation results of Watershed method, initially it construct a RAG(Region Merging Graph) for optimal RAT by applying the most popular MRF(Markov Random Field) method . Consequently, the inter-region comparison is presented by applying the watershed method in Spatial Domain and Fuzzy C-Means clustering method in Feature Space for image mapping to compute the Optimal Region Amalgamation. Further, to determine the Feature space and domain space illustration of the brain MR image segmentation, the SGD (Spatial Graph Depiction) is presented that is computed with FSD (Feature Space Depiction) which is obtained by watershed partitioning and FCM clustering method. The experimental results on multimodal brain MR image datasets presents that the proposed novel Optimal Region Amalgamation Technique (RAT) exhibits more promising MR images segmentation results with compared to the traditional watershed method. Finally, an assessment and evaluation of the state-of-the-art brain tumor segmentation methods are presented and future directions to improve and standardize the detection and segmentation of brain tumor into daily clinical treatment are addressed.

Markov Random Fields Integrating Adaptive Interclass-Pair Penalty and Spectral Similarity for Hyperspectral Image Classification

This paper presents a novel Markov random field (MRF) method integrating adaptive interclass-pair penalty (aICP2) and spectral similarity information (SSI) for hyper-spectral image (HSI) classification. aICP2 structurally combines $K(K - 1)/2$ (K is the number of classes) classical “Potts model” with $K(K - 1)/2$ interaction coefficients. aICP2 tries a new way to solve the key problems, insufficient correction within homogeneous regions, and over-smoothness at class boundaries, in MRF-based HSI classification. It is assumed that different class pairs should be assigned with various degrees of penalties in MRF smoothness process, according to pairwise class separability and spatial class confusion in raw classification map. The Fisher ratio is modified to measure pairwise class separability with a training set. And, gray level co-occurrence matrix is used to measure spatial class confusion degree. Then, aICP2 is constructed by combining Fisher ratio and GCLM. aICP2 applies larger penalty on class pairs that confuse with each other seriously to provide sufficient smoothness, and vice versa. In addition, to protect class edges and details, SSI is introduced to make the penalty of related neighboring pixels small. aICP2ssi denotes the integration of aICP2 and SSI. The further improved method is both interclass-pair and interpixel adaptive in spatial term. A graph-cut-based $\alpha - \beta $ -swap method is introduced to optimize the proposed energy function. The experimental results on real HSI data indicate that the proposed method outperforms compared MRF-based and other spectral–spatial approaches in terms of classification accuracies and region uniformity.

Segmentation of sonar imagery using convolutional neural networks and Markov random field

In this paper, we present a novel method incorporating convolutional neural networks (CNN) into Markov random field (MRF) to automatically segment side scan sonar (SSS) images into object-highlight, object-shadow and sea-bottom reverberation areas. As a widely used ocean survey sensor, SSS provides high-resolution maps of the seafloor. Automatically segmenting SSS in real time can assist the navigation and path-planning of autonomous underwater vehicles. However, for the speckle noise and intensity inhomogeneity in the SSS images, it is difficult to find a robust SSS segmentation method. These facts motivate us to explore efficient CNN architectures to solve these problems. For pixel-level SSS segmentation, to use the context information and the details around a central pixel simultaneously, the CNN with multi-scale inputs (MSCNN) is employed. Besides, to mitigate the impact of the class imbalance problem, two MSCNN training strategies are introduced, which are based on data augmentation and ensemble learning. Furthermore, to take into account the local dependencies of class labels, the results of MSCNN are used to initialize MRF to get the final segmentation maps. Experimental results on real SSS images reveal that the proposed segmentation method outperforms MRF, CNN and semantic segmentation methods such as fully convolutional network and Segnet in segmentation accuracy and generalization performance. Moreover, the efficiency of the proposed method is proved on retinal image dataset.

Segmentation for remote-sensing imagery using the object-based Gaussian-Markov random field model with region coefficients

ABSTRACTThe Markov random field (MRF) model is a widely used method for remote-sensing image segmentation, especially the object-based MRF (OMRF) method has attracted great attention in recent years. However, the OMRF method usually fails to capture the correlation between regional features by just considering the mixed-Gaussian model. In order to solve this problem and improve the segmentation accuracy, this article proposes a new method, object-based Gaussian-Markov random field model with region coefficients (OGMRF-RC), for remote-sensing image segmentation. First, to describe the complicated interactions among regional features, the OGMRF-RC method employs the region size and edge information as region coefficients to build the object-based linear regression equation (OLRE) for each region. Second, the classic Gaussian-Markov model is extended to region level for modelling the errors in OLREs. Finally, the segmentation is achieved through a principled probabilistic inference designed for the OGMRF-RC method. Experimental results over synthetic texture images and remote-sensing images from different datasets show that the proposed OGMRF-RC method can achieve more accurate segmentation than other state-of-the-art MRF-based methods and the method using convolutional neural networks.

Efficient Confidence-Based Hierarchical Stereo Disparity Upsampling for Noisy Inputs

Disparity upsampling techniques aim to restore high-resolution disparity maps from lowresolution disparity inputs. These inputs must be of high quality and are often obtained via complicated passive or active 3D reconstruction methods. Each pixel in the input disparity maps guides the disparity assignment in the upsampling process. The quality of the upsampled results will decrease if the initial disparity inputs are noisy, as the upsampled results are closely related to the initial inputs.We herein propose a hierarchical confidence-based upsampling framework that can be used to obtain relatively high quality upsampled results even under the noisy inputs. Specifically designed confidence measuring schemes are employed in our upsampling process, allowing the disparity assignment of only high-confidence pixels. For an effective depth quality evaluation, we present a novel classification of the confidence according to depth- and texture-related information and develop a confidence examination method with improved precision by combining multiple depth confidence evaluation methods. Our hierarchical pipeline contains 3 steps: confidence-based upsampling, confidence-based fine-tuning and confidence-based optimization. The upsampling combines multichannel information. Fine-tuning is carried out using the stereo texture information. Optimization is conducted utilizing the Markov random field method. All these proposed methods work together to suppress the low-confidence pixels and propagate the high-confidence pixels in the upsampling process. The cumulative error distribution is further analyzed, revealing the effectiveness of our confidence evaluation. Extensive comparison experiments are also performed using both the ground truth and stereo matching disparity maps as inputs to demonstrate the advantage of our framework over state-of-the-art upsampling methods.

Sparse-Based Classification of Hyperspectral Images Using Extended Hidden Markov Random Fields

This paper proposes a spectral–spatial method for the classification of hyperspectral images (HSIs). The proposed method consists of suggesting two main techniques. First, to construct dictionaries and sparse codes containing the least spatial–spectral correlation, an objective function for a sparse classifier, namely optimum dictionary learning (ODL), is proposed. This objective function combines the spectral and spatial features in a hierarchical fashion. ODL is applied in the nonsubsampled pyramid domain to exploit the multiresolution property of pyramids. Second, to enhance the classification accuracy, an extended hidden Markov random field (EHMRF) is suggested. The EHMRF uses the Poisson distribution and weighted mean regularization to overcome Poisson noise and to increase the local neighborhood energy consistency, respectively. Finally, in order to estimate the parameter of the Poisson distribution, an iterative maximum likelihood (ML) algorithm is used. In comparison with some of state-of-the-art hidden Markov random field and sparse-based classifier methods, the experimental results of the proposed method (ODL-EHMRF-ML) show that it significantly increases the classification accuracy of HSIs.

Precise Markov random field model-based phase unwrapping method for airborne interferometric synthetic aperture radar imaging

Markovian model approach with graph cuts is effectively applied for phase unwrapping (PU) in interferometric synthetic aperture radar (InSAR) domain particularly for data with big noises and discontinuities. The first-order prior model in Markov random field (MRF) is widely used in Markovian model approach, which is able to solve PU problem in homogeneous scenarios. However, the first-order MRF method cannot interpret local details well in abundant patterns of interferograms under the requirement of high accuracy due to the limitation of first-order prior information. Hence, the conventional first-order model is no longer applicable in this situation. In terms of this problem, the shortcoming of the first-order model is analyzed, and a precise second-order MRF approach is proposed in this paper. Furthermore, a graph cuts-based optimization PU algorithm is designed under a high-order MRF mode, which improves the PU accuracy in detail areas. Without apparent increase of computational complexity, the proposed method is more robust compared with the conventional first-order MRF-based methods. A set of experimental results on both simulation and real measured millimeter-wave InSAR data illustrate the effectiveness of the proposal.

Low-Power Noise-Immune Nanoscale Circuit Design Using Coding-Based Partial MRF Method

Reliability is one of the major concerns for ultralow power circuit designs. Markov random field (MRF) techniques have been applied to logic circuits to resist random noise when operating under ultralow supply voltage or sub-threshold voltage. Although conventional MRF networks can be easily mapped onto simple logic circuits, it becomes difficult when the circuits are large and complex. In this paper, we present a general coding-based partial MRF (CPMRF) method for multi-logic operations in one basic unit, which is referred to as a CPMRF pair. A CPMRF pair saves circuit area by sharing a common MRF network. It also inherits noise immunity from the MRF theory while obtaining noise immunity from the coding structure as a combination of robust “1s” and “0s.” The resulting architectures become more cost effective than conventional ones. To validate the performance of our proof-of-concept design, we fabricated a carry-lookahead adder implemented by the proposed CPMRF pairs using IBM 130-nm CMOS technology. Measurement results indicate that the CPMRF CLA can achieve high noise tolerance with 20% improvement while occupying 37.7% less area and reducing power consumption by 93% compared with the master-and-slave MRF CLA design.

Landslide Inventory Mapping From Bitemporal High-Resolution Remote Sensing Images Using Change Detection and Multiscale Segmentation

Landslide inventory mapping (LIM) plays an important role in hazard assessment and hazard relief. Even though much research has taken place in past decades, there is space for improvements in accuracy and the usability of mapping systems. In this paper, a new landslide inventory mapping framework is proposed based on the integration of the majority voting method and the multiscale segmentation of a postevent images, making use of spatial feature of landslide. Compared with some similar state-of-the-art methods, the proposed framework has three advantages: 1) the generation of LIM is almost automatic; 2) the framework can achieve more accurate results because it takes into account the landslide spatial information in an irregular manner through multisegmentation of the postevent image and object-based majority voting (MV); and 3) it needs less parameter tuning. The proposed framework was applied to four landslide sites on Lantau Island, Hong Kong. Compared with existing methods, including region level set evolution (RLSE), edge level set evolution (ELSE) and change detection Markov random field (CDMRF) methods, quantitative evaluation shows the proposed framework is competitive in terms of Completeness. The framework outperformed RLSE, ELSE, and CDMRF for the four experiments by more than 9% in Correctness and by 8% in Quality. To the authors' knowledge, this is the first-time that landslide spatial information has been utilized through the integration of multiscale segmentation of postevent image with the MV approach to obtain LIM using high spatial resolution remote sensing images. The approach is also of wide generality and applicable to other kinds of land cover change detection using remote sensing images.

LANDSLIDE INVENTORY MAPPING FROM BITEMPORAL 10 m SENTINEL-2 IMAGES USING CHANGE DETECTION BASED MARKOV RANDOM FIELD

Abstract. Landslide inventory mapping is essential for hazard assessment and mitigation. In most previous studies, landslide mapping was achieved by visual interpretation of aerial photos and remote sensing images. However, such method is labor-intensive and time-consuming, especially over large areas. Although a number of semi-automatic landslide mapping methods have been proposed over the past few years, limitations remain in terms of their applicability over different study areas and data, and there is large room for improvement in terms of the accuracy and automation degree. For these reasons, we developed a change detection-based Markov Random Field (CDMRF) method for landslide inventory mapping. The proposed method mainly includes two steps: 1) change detection-based multi-threshold for training samples generation and 2) MRF for landslide inventory mapping. Compared with the previous methods, the proposed method in this study has three advantages: 1) it combines multiple image difference techniques with multi-threshold method to generate reliable training samples; 2) it takes the spectral characteristics of landslides into account; and 3) it is highly automatic with little parameter tuning. The proposed method was applied for regional landslides mapping from 10 m Sentinel-2 images in Western China. Results corroborated the effectiveness and applicability of the proposed method especially the capability of rapid landslide mapping. Some directions for future research are offered. This study to our knowledge is the first attempt to map landslides from free and medium resolution satellite (i.e., Sentinel-2) images in China.

SAR IMAGE CHANGE DETECTION BASED ON FUZZY MARKOV RANDOM FIELD MODEL

Abstract. Most existing SAR image change detection algorithms only consider single pixel information of different images, and not consider the spatial dependencies of image pixels. So the change detection results are susceptible to image noise, and the detection effect is not ideal. Markov Random Field (MRF) can make full use of the spatial dependence of image pixels and improve detection accuracy. When segmenting the difference image, different categories of regions have a high degree of similarity at the junction of them. It is difficult to clearly distinguish the labels of the pixels near the boundaries of the judgment area. In the traditional MRF method, each pixel is given a hard label during iteration. So MRF is a hard decision in the process, and it will cause loss of information. This paper applies the combination of fuzzy theory and MRF to the change detection of SAR images. The experimental results show that the proposed method has better detection effect than the traditional MRF method.

Application of Markov Random Field on Microwave Near-Field Imaging of Breast Tumor

This paper solved an inverse scattering problem of near field under some conditions and presented a novel microwave near-field imaging method of breast tissues with tumors, which is based on the contrast of dielectric properties between normal breast tissues and tumors. We used microwave scatter- sounding method to detect the breast and obtained the reconstruction projection data. Then modeled the dielectric properties distribution of breast tissues with Markov Random Field (MRF) method. With the characteristics of radial distribution of the data, we established the relationship between dielectric properties distribution and detected data. Using Maximum a Posteriori (MAP) estimation, we obtained an optimization problem and solved it. The simulation result shows that the method is effective for imaging.

Synthetic Aperture Radar Image Recognition Using Contour Features

Synthetic aperture radar (SAR) images play an important role in nowadays target recognition. However, due to the intrinsic impact of speckle noise in SAR images, many details of the targets are confused, thus causing difficulty on target recognition. Therefore, this paper introduces a novel image recognition method which performs well on optical images. In our paper, the target is first extracted by combining the multi-resolution Markov random field method with the wavelet transform. Such a process is able to reduce the speckle noise from different resolutions. Then, morphological filtering is utilized to further reduce the remaining noise as well as smooth the boundary. With the smoothed target, we adopt the state-ofthe- art shape representation method called bag of contour fragments for target classification, which encodes the target into different contours. Then, by using support vector machine (SVM), the contours of each target can be well classified and the target recognition can be achieved. Finally, the experimental results validate the superior performance of our method on the recognition accuracy.

A Markov random field integrating spectral dissimilarity and class co-occurrence dependency for remote sensing image classification optimization

This paper develops a novel Markov Random Field (MRF) model for edge-preserving spatial regularization of classification maps. MRF methods based on the uniform smoothness lead to oversmoothed solutions. In contrast, MRF methods which take care of local spectral or gradient discontinuities, lead to unexpected object particles around boundaries. To solve these key problems, our developed MRF method first establishes a spatial energy function integrating local spectral dissimilarity to smooth the initial classification map while preserving object boundaries. Second, a new anisotropic spatial energy function integrating the class co-occurrence dependency is constructed to regularize pixels around object boundaries. The effectiveness of the method is tested using a series of remote sensing data sets. The obtained results indicate that the method can avoid oversmoothing and significantly improve the classification accuracy with regards to traditional MRF classification models and some other state-of-the-art methods.

Automatic Choroidal Layer Segmentation Using Markov Random Field and Level Set Method.

The choroid is an important vascular layer that supplies oxygen and nourishment to the retina. The changes in thickness of the choroid have been hypothesized to relate to a number of retinal diseases in the pathophysiology. In this paper, an automatic method is proposed for segmenting the choroidal layer from macular images by using the level set framework. The three-dimensional nonlinear anisotropic diffusion filter is used to remove all the optical coherence tomography (OCT) imaging artifacts including the speckle noise and to enhance the contrast. The distance regularization and edge constraint terms are embedded into the level set method to avoid the irregular and small regions and keep information about the boundary between the choroid and sclera. Besides, the Markov random field method models the region term into the framework by correlating the single-pixel likelihood function with neighborhood information to compensate for the inhomogeneous texture and avoid the leakage due to the shadows cast by the blood vessels during imaging process. The effectiveness of this method is demonstrated by comparing against other segmentation methods on a dataset with manually labeled ground truth. The results show that our method can successfully and accurately estimate the posterior choroidal boundary.

Object-based change detection method using refined Markov random field

In order to fully consider the local spatial constraints between neighboring objects in object-based change detection (OBCD), an OBCD approach is presented by introducing a refined Markov random field (MRF). First, two periods of images are stacked and segmented to produce image objects. Second, object spectral and textual histogram features are extracted and G-statistic is implemented to measure the distance among different histogram distributions. Meanwhile, object heterogeneity is calculated by combining spectral and textual histogram distance using adaptive weight. Third, an expectation-maximization algorithm is applied for determining the change category of each object and the initial change map is then generated. Finally, a refined change map is produced by employing the proposed refined object-based MRF method. Three experiments were conducted and compared with some state-of-the-art unsupervised OBCD methods to evaluate the effectiveness of the proposed method. Experimental results demonstrate that the proposed method obtains the highest accuracy among the methods used in this paper, which confirms its validness and effectiveness in OBCD.

Detection of Melanoma Skin Cancer in Dermoscopy Images

Malignant melanoma is the most hazardous type of human skin cancer and its incidence has been rapidly increasing. Early detection of malignant melanoma in dermoscopy images is very important and critical, since its detection in the early stage can be helpful to cure it. Computer Aided Diagnosis systems can be very helpful to facilitate the early detection of cancers for dermatologists. In this paper, we present a novel method for the detection of melanoma skin cancer. To detect the hair and several noises from images, pre-processing step is carried out by applying a bank of directional filters. And therefore, Image inpainting method is implemented to fill in the unknown regions. Fuzzy C-Means and Markov Random Field methods are used to delineate the border of the lesion area in the images. The method was evaluated on a dataset of 200 dermoscopic images, and superior results were produced compared to alternative methods.

Interactive Image Segmentation using Improved Adaptive Markov Random Field Approach

Background/Objectives: To interactively split an object of interest from the remaining image with better smoothing by introducing improved adaptive Markov Random field resolving the problem of much noise. Methods/Statistical analysis: By utilizing a Dirichlet process multiple - view learning scheme, the unlabelled pixel labels are calculated by using the seed pixels. It is used for supporting the multiple-view learning in order to incorporate the constraints of both appearance and boundary constraint, and the Dirichlet process mixture-based nonlinear classification for concurrently modelling the image features and distinguishing the differences between the classes of object and background. The MRF field is utilized for providing the smoothness in segment labels. Findings: In Markov Random Field (MRF) based scheme, only the pixel and the surrounding pixels relationships are considered. The microscopic image processing result along with low noise is bad. To solve this problem, the adaptive MRF method based on region; it exploits Graph Cuts for inference. The different type of images produces the different pre-segmentation results. The connection degree between current and its linked regional blocks is denoted by connection parameter. If the connection parameter value is high, large connection degree between regional block and neighbouring regional blocks is defined. Otherwise lower connection degree between regional block and neighbourhood regional blocks. The adaptive MRF smoothing is more accurate and efficient segmentation result than the traditional MRF based Smoothing method. However, the appropriate parameter selection is a difficult task for practical image segmentation which can be solved by introducing an improved adaptive MRF model by using a modified graph cutter. Thus, the image segmentation is achieved based on modified graph-cut model using a novel energy function without the regularizing parameter. Improvements/Applications: The improved adaptive Markov Random Field approach interactively segment images with better smoothness than most of the current approaches.