Iterated Conditional Modes Research Articles

MAP-Based Motion Refinement Algorithm for Block-Based Motion-Compensated Frame Interpolation

A new motion vector field (MVF) refinement algorithm is proposed for block-based motion-compensated frame interpolation. Under the assumption that an observed MVF, such as the result of a block-based motion estimation (BME), is a degraded version of the true MVF, the true MVF is estimated from the observation through maximum a posteriori probability (MAP) estimation. To define the posterior probability of the true MVF, the degradation is modeled as a locally stationary additive Gaussian noise, so the variance of the noise represents the unreliability of the observed motion vector (MV). The noise variance is directly estimated from the observation vector and its select neighbors. The prior distribution of the true MVF is designed to rely on the distances between the MV and its neighbors and to properly smooth the false MVs in the observation. The MAP estimate of the true MVF is obtained via the iterative conditional mode method. The outcome is a set of iterative update equations, which produce the kth estimate of the true MV of a block by combining, according to the estimated noise variance, the observation and the neighboring (k - 1)th estimates. Experimental results prove that the proposed algorithm achieves performances comparable with those of several existing MAP-based BME algorithms at a much lower computational complexity.

Segmentation of Moving Object Using Background Subtraction Method in Complex Environments

Background subtraction is an extensively used approach to localize the moving object in a video se- quence. However, detecting an object under the spatiotem- poral behavior of background such as rippling of water, moving curtain and illumination change or low resolution is not a straightforward task. To deal with the above-men- tioned problem, we address a background maintenance scheme based on the updating of background pixels by estimating the current spatial variance along the temporal line. The work is focused to immune the variation of local motion in the background. Finally, the most suitable label assignment to the motion field is estimated and optimized by using iterated conditional mode (ICM) under a Mar- kovian frame- work. Performance evaluation and compari- sons with the other well-known background subtraction methods show that the proposed method is unaffected by the problem of aperture distortion, ghost image, and high frequency noise.

Comparing support vector machine contextual approaches for urban area classification

ABSTRACTSupport vector machine (SVM) has been receiving a great deal of attention for remote sensing data classification. Although the original formulation of this method does not incorporate contextual information, lately different formulations have been proposed to incorporate such information, with the aim of improving the mapping accuracy. In general, these proposals modify the SVM training phase or integrate the SVM classifications in stochastic models. Recently, two new contextual versions of SVM, context adaptive and competitive translative SVM (CaSVM and CtSVM, respectively), were proposed in literature. In this work, two case studies of urban area classification, using IKONOS-II and hyperspectral digital imagery collection experiment (HYDICE) data sets were conducted to compare SVM, SVM integrated with the iterated conditional modes (ICM) stochastic algorithm, SVM smoothed using the mode filter and the recent approaches CaSVM and CtSVM. The results indicated that although it possesses a high computational cost, the CaSVM method was able to produce classification results with similar accuracy (using kappa coefficient) to those obtained using SVM integrated with ICM (SVM + ICM) and the mode filter (SVM + Mode), all of them found statistically superior to the SVM result at 95% confidence level for the IKONOS-II image. For HYDICE image, all results were found statistically insignificant at 95% confidence level. Investigation of what happens at transition regions between classes, however, showed that some methods can present superior performance. To this objective, a new performance measure, called upsilon coefficient, was introduced in this work, which measures the impact that the smoothing effect, typical of contextual methods, can have in distorting the edges between regions. With this new measure was found that CaSVM is the one which has better performance followed with SVM + ICM.

Efficient, Edge-Aware, Combined Color Quantization and Dithering.

In this paper, we present a novel algorithm to simultaneously accomplish color quantization and dithering of images. This is achieved by minimizing a perception-based cost function, which considers pixel-wise differences between filtered versions of the quantized image and the input image. We use edge aware filters in defining the cost function to avoid mixing colors on the opposite sides of an edge. The importance of each pixel is weighted according to its saliency. To rapidly minimize the cost function, we use a modified multi-scale iterative conditional mode (ICM) algorithm, which updates one pixel a time while keeping other pixels unchanged. As ICM is a local method, careful initialization is required to prevent termination at a local minimum far from the global one. To address this problem, we initialize ICM with a palette generated by a modified median-cut method. Compared with previous approaches, our method can produce high-quality results with a fewer visual artifacts but also requires significantly less computational effort.

A new image segmentation framework based on two-dimensional hidden Markov models

Image segmentation is one of the fundamental problems in computer vision. In this work, we present a new seg- mentation algorithm that is based on the theory of two-dimensional hidden Markov models (2D-HMM). Unlike most 2D-HMM approaches we do not apply the Viterbi algorithm; instead we present a computationally efficient algorithm that propagates the state probabilities through the image. Our algorithm, called Complete Enumeration Iteration (CEP), is flexible in the sense that it allows the use of different probability distributions as emission probabilities. Not only do we compare the performance of dif- ferent probability functions plugged into our framework but also propose three methods to update the distributions of each state online during the segmentation process. We compare our algorithm with a 2D-HMM standard algorithm and Iterated Condi- tional Modes (ICM) using real world images like a radiography or a satellite image as well as synthetic images. The experimental results are evaluated by the kappa coefficient ( κ). In those cases where the average κ coefficient is higher than 0.7 we observe an average relative improvement of 8% of CEP with respect to the benchmark algorithms. For all other segmentation tasks CEP shows no significant improvement. Besides that, we demonstrate how the choice of the emission probability can have great in- fluence on the segmentation results. Surprisingly, we observe that the normal distribution is an appropriate density function for many segmentation tasks.

OBJECT-ORIENTED CHANGE DETECTION BASED ON SPATIOTEMPORAL RELATIONSHIP IN MULTITEMPORAL REMOTE-SENSING IMAGES

Abstract. In this paper a novel object-oriented change detection approach in multitemporal remote-sensing images is proposed. In order to improve post classification comparison (PCC) performance, we propose to exploit spatiotemporal relationship between two images acquired at two different times. The probabilities of class transitions are used to describe the temporal dependence information, while the Markov Random Field (MRF) model is utilized to represent the spatial-contextual information. Training sets are required to get initial classification results b maximum likelihood method (ML). Then an estimation procedure: iterated conditional mode (ICM) is present to revise the classification results. Change detection (change/no change) and change type recognitions (from-to types of change) are achieved by compare classification maps acquired at two different times. Experimental results on two QuickBird images confirm that the proposee method can provide higher accuracy than the PCC method, which ignores spatiotemporal relationship between two images.

SAR-based change detection using hypothesis testing and Markov random field modelling

Abstract. The objective of this study is to automatically detect changed areas caused by natural disasters from bi-temporal co-registered and calibrated TerraSAR-X data. The technique in this paper consists of two steps: Firstly, an automatic coarse detection step is applied based on a statistical hypothesis test for initializing the classification. The original analytical formula as proposed in the constant false alarm rate (CFAR) edge detector is reviewed and rewritten in a compact form of the incomplete beta function, which is a builtin routine in commercial scientific software such as MATLAB and IDL. Secondly, a post-classification step is introduced to optimize the noisy classification result in the previous step. Generally, an optimization problem can be formulated as a Markov random field (MRF) on which the quality of a classification is measured by an energy function. The optimal classification based on the MRF is related to the lowest energy value. Previous studies provide methods for the optimization problem using MRFs, such as the iterated conditional modes (ICM) algorithm. Recently, a novel algorithm was presented based on graph-cut theory. This method transforms a MRF to an equivalent graph and solves the optimization problem by a max-flow/min-cut algorithm on the graph. In this study this graph-cut algorithm is applied iteratively to improve the coarse classification. At each iteration the parameters of the energy function for the current classification are set by the logarithmic probability density function (PDF). The relevant parameters are estimated by the method of logarithmic cumulants (MoLC). Experiments are performed using two flood events in Germany and Australia in 2011 and a forest fire on La Palma in 2009 using pre- and post-event TerraSAR-X data. The results show convincing coarse classifications and considerable improvement by the graph-cut post-classification step.

Improving the runtime of MRF based method for MRI brain segmentation

Image segmentation is one of the important parts in medical image analysis. Markov random field (MRF) is one of the successful methods for MRI image segmentation, but conventional MRF methods suffer from high computational cost. MRI images have high level of artifacts such as Partial Volume Effect (PVE), intensity non uniformity (INU) and other noises, so using global optimization methods like simulated annealing (SA) for optimization step is more appropriate than other local optimization methods such as Iterative Conditional Modes (ICM). On the other hand, these methods also has heavy computational burden and they are not appropriate for real time task. This paper uses a proper combination of clustering methods and MRF and proposes a preprocessing step for MRF method for decreasing the computational burden of MRF for segmentation. The results show that the preprocessing step increased the speed of segmentation algorithm by a factor of about 10 and have no large impact on the accuracy of segmentation. Moreover, different clustering methods can be used for the first step and estimation of the parameters. Therefore, using of powerful clustering methods can provide a better segmentation results.

Selecting massive variables using an iterated conditional modes/medians algorithm

Empirical Bayes methods are designed in selecting massive variables, which may be inter-connected following certain hierarchical structures, because of three attributes: taking prior information on model parameters, allowing data-driven hyperparameter values, and free of tuning parameters. We propose an iterated conditional modes/medians (ICM/M) algorithm to implement empirical Bayes selection of massive variables, while incorporating sparsity or more complicated a priori information. The iterative conditional modes are employed to obtain data-driven estimates of hyperparameters, and the iterative conditional medians are used to estimate the model coefficients and therefore enable the selection of massive variables. The ICM/M algorithm is computationally fast, and can easily extend the empirical Bayes thresholding, which is adaptive to parameter sparsity, to complex data. Empirical studies suggest competitive performance of the proposed method, even in the simple case of selecting massive regression predictors.

Synthetic aperture radar image segmentation using fuzzy label field‐based triplet Markov fields model

The recently proposed triplet Markov random fields (TMF) model is very suitable for dealing with non-stationary image segmentation. However, influenced by multiplicative speckle noise, synthetic aperture radar image (SAR) is dim and blurred in the boundaries of different areas, making it difficult to locate boundary accurately in the segmentation process. Thus, in this study, the authors propose a new segmentation algorithm using fuzzy label field-based TMF model for SAR images. In the proposed algorithm, the value of each site in the label field is extended from a finite discrete set in the classical TMF model to a continuous one, in order to describe the memberships of each pixel to different classes. A fuzzy energy function is constructed to describe the joint prior distribution of the fuzzy label field and the auxiliary field. The construction of fuzzy energy function also takes into account four direction information and degree of difference between neighbouring pixels. Iterative conditional estimation method and maximum posterior mode criterion are applied to implement parameter estimation and segmentation. Experimental results on simulated data and real SAR images demonstrate the effectiveness of the proposed algorithm.

Improving SAR-Based Urban Change Detection by Combining MAP-MRF Classifier and Nonlocal Means Similarity Weights

In remote sensing change detection, Markov random field (MRF) has been used successfully to model the prior probability using class-labels dependencies. MRF has played an important role in the detection of complex urban changes using optical images. However, the preservation of details in urban change analysis turns out to be a highly complex task if multitemporal SAR images with their speckle are to be used. Here, the ability of MRF to preserve geometric details and to combat speckle effect at the same time becomes questionable. Blob-region phenomenon and fine structures removal are common consequences of the application of traditional MRF-based change detection algorithm. To overcome these limitations, the iterated conditional modes (ICM) framework for the optimization of the maximum a posteriori (MAP-MRF) criterion function is extended to include a nonlocal probability maximization step. This probability model, which characterizes the relationship between pixels' class-labels in a nonlocal scale, has the potential to preserve spatial details and to reduce speckle effects. Two multitemporal SAR datasets were used to assess the proposed algorithm. Experimental results using three density functions [i.e., the log normal (LN), generalized Gaussian (GG), and normal distributions (ND)] have demonstrated the efficiency of the proposed approach in terms of detail preservation and noise suppression. Compared with the traditional MRF algorithm, the proposed approach proved to be less-sensitive to the value of the contextual parameter and the chosen density function. The proposed approach has also shown less sensitivity to the quality of the initial change map when compared with the ICM algorithm.

Automatic Registration between High Resolution SAR Image and GIS Road Network Based on Pairwise Constraint and ICM Optimization

Synthetic aperture radar (SAR) technique has been widely applied in disaster monitoring and evaluation, because of its all-weather and all-time capability. In particular, the integration of high resolution SAR image and GIS data has been proved to be an effective solution to change detection for rapid disaster response. However, accurate registration of SAR image and GIS data remains a challenging task. On the one hand, SAR image and GIS data are two different representation levels of the real world. GIS data lacks of grey value and texture information, thus traditional use of grey-based and texture-based feature descriptors for feature matching in GIS data is impossible. On the other hand, the extraction of matching primitives is difficult in the SAR image due to the speckle noise. Moreover, though features can be extracted in the SAR image, the matching can be difficult since features may be fractured and missing. In this paper, we propose a new method for vector-to-SAR image registration which utilizes conjugate line-pair intersections as matching primitives. The core idea consists in two aspects: 1) Different from traditional road intersection based registration methods, conjugate line-pair intersections are employed as matching primitive in the proposed method. 2) To find out the optimal set of matching features, a matching technique that using pairwise constraint and the iterative conditional mode (ICM) optimization algorithm is introduced. Experiment results proved the reliability and feasibility of the proposed method.

A Markov random field based approach to the identification of meat and bone meal in feed by near-infrared spectroscopic imaging

Contaminated meat and bone meal (MBM) in animal feedstuff has been the source of bovine spongiform encephalopathy (BSE) disease in cattle, leading to a ban in its use, so methods for its detection are essential. In this study, five pure feed and five pure MBM samples were used to prepare two sets of sample arrangements: set A for investigating the discrimination of individual feed/MBM particles and set B for larger numbers of overlapping particles. The two sets were used to test a Markov random field (MRF)-based approach. A Fourier transform infrared (FT-IR) imaging system was used for data acquisition. The spatial resolution of the near-infrared (NIR) spectroscopic image was 25 μm × 25 μm. Each spectrum was the average of 16 scans across the wavenumber range 7,000-4,000 cm(-1), at intervals of 8 cm(-1). This study introduces an innovative approach to analyzing NIR spectroscopic images: an MRF-based approach has been developed using the iterated conditional mode (ICM) algorithm, integrating initial labeling-derived results from support vector machine discriminant analysis (SVMDA) and observation data derived from the results of principal component analysis (PCA). The results showed that MBM covered by feed could be successfully recognized with an overall accuracy of 86.59% and a Kappa coefficient of 0.68. Compared with conventional methods, the MRF-based approach is capable of extracting spectral information combined with spatial information from NIR spectroscopic images. This new approach enhances the identification of MBM using NIR spectroscopic imaging.

Segmentation of cone-beam CT using a hidden Markov random field with informative priors

Cone-beam computed tomography (CBCT) has enormous potential to improve the accuracy of treatment delivery in image-guided radiotherapy (IGRT). To assist radiotherapists in interpreting these images, we use a Bayesian statistical model to label each voxel according to its tissue type. The rich sources of prior information in IGRT are incorporated into a hidden Markov random field model of the 3D image lattice. Tissue densities in the reference CT scan are estimated using inverse regression and then rescaled to approximate the corresponding CBCT intensity values. The treatment planning contours are combined with published studies of physiological variability to produce a spatial prior distribution for changes in the size, shape and position of the tumour volume and organs at risk. The voxel labels are estimated using iterated conditional modes. The accuracy of the method has been evaluated using 27 CBCT scans of an electron density phantom. The mean voxel-wise misclassification rate was 6.2%, with Dice similarity coefficient of 0.73 for liver, muscle, breast and adipose tissue. By incorporating prior information, we are able to successfully segment CBCT images. This could be a viable approach for automated, online image analysis in radiotherapy.

Remote sensing classification method based on image segment spatial relationship

An image classification method based on the spatial relationship of image segment is proposed with the purpose of excavating the spatial relationship between image segments and compensating for deficiencies of the traditional image classification method based on spectral information.Image segmentation is used to get image segments for original image classification employing maximum likelihood(ML) method.Then the spatial relationship of image segments is described by Markov random field(MRF).Quantitative spatial relationship can be obtained by class adjacency matrix(CAM) so as to revise the result of classification.After that the iterated conditional mode(ICM) algorism for classification is presented,which can yield results with higher accuracy.Experimental results show that this method has been functioning well in classification experiments with high resolution remote sensing images.

Influence of Clique Potential Parameters on Classification Using Bayesian MRF Model for Remote Sensing Image in Dali Erhai Basin

Image classification of remote sensing data is an important topic and long-term tasks in applications [1]. Markov random field (MRF) has more advantages in processing contextual information [2]. Bayesian approach enables the incorporation of prior model and likelihood distribution, this paper has formulated a Bayesian-MRF classification model based on MAP-ICM framework. It uses Potts model in label field and assume Gaussian distribution in observation field. According to maximum a posteriori (MAP) criterion, each new classified label can be obtained by the minimum of energy using Iterated Conditional Modes (ICM) algorithm. Finally, classification tasks are carried out by Bayesian-MRF classification model. Experimental results show that: (1) Clique potential parameters affect classification greatly. When it is 0.5, the classification accuracy reaches maximum with the best classification result for study area of Dali Erhai Lake basin using landsat TM data. (2) Bayesian MRF model have obvious advantages in classification for neighbourhood pixels so that it can separate Shadow class from Water class because the Shadow in mountain areas is very similar to Water in spectrum. In this case study, the best classification accuracy reaches 95.8%. The approaches and results will have important reference value for applications such as land use/cover classification, environment/ecological monitoring etc.

Top-down Inference with Relabeling and Mapping Rules in Hierarchical MRF for Image Segmentation

摘要: 针对彩色图像分割问题,研究Markov 随机场(Markov random fields, MRF)模型内迭代条件模式(Iterative conditional mode, ICM)方法的标记推理策略. 通过小波分解构造图像多尺度表达,针对顶层图像先验标记获取问题,改进原始谱聚类算法, 通过近邻传播自动确定图像的聚类参数,运用集成学习提高算法的稳定性和准确度. 对其他各尺度图像,通过分析尺度关联下的区域特征变化,结合不同尺度间的特征相似性和同一尺度内空间邻域的一致性, 提出一种立体结构描述下的尺度--空间映射法则.通过定量和定性的分割实验,结果表明本文算法具有良好的准确性、鲁棒性和普适性. 关键词: 层次Markov随机场 / 集成标记 / 层间映射推理 / 图像分割

Infrared dim small target detection algorithm based on generative Markov random field and local statistic characteristic

Dim small target detection problem in infrared complex background was formulated as a binary classification problem of background and target in the theoretical framew ork of Markov random field( MRF). Based on the posterior probability model of MRF,a method using prior information of target SCR( signal-to-clutter ratio) and local statistic characteristic of infrared image w as proposed to construct the posterior probability model of observed image. The classic iterated conditional mode( ICM) w as used to estimate the optimal labeling image. Simulation and experimental results show that the proposed algorithm effectively reduces the false labeling probability of background,w hile maintaining a high probability of correct labeling of target. In addition,for using image's local statistic characteristic in modeling,the proposed algorithm also reduces the correlation betw een labeled results and model parameters w hich contributes to improvement on the convergence speed of estimating the optimal labeling.

Gentle ICM energy minimization for Markov random fields with smoothness-based priors

Coordinate descent, also known as iterated conditional mode (ICM) algorithm, is a simple approach for minimizing the energy defined by a Markov random field. Unfortunately, the ICM is very sensitive to the initial values and usually only finds a poor local minimum of the energy. A few modifications of the ICM algorithm are discussed here that ensure a more `gentle' descent during the first iterations of the algorithm and that lead to substantial performance improvements. It is demonstrated that the modified ICM can be competitive to other optimization algorithms on a set of vision problems such as stereo depth estimation, image segmentation, image denoising and inpainting.

Cloud image detection based on Markov Random Field

In order to overcome the disadvantages of low accuracy rate, high complexity and poor robustness to image noise in many traditional algorithms of cloud image detection, this paper proposed a novel algorithm on the basis of Markov Random Field (MRF) modeling. This paper first defined algorithm model and derived the core factors affecting the performance of the algorithm, and then, the solving of this algorithm was obtained by the use of Belief Propagation (BP) algorithm and Iterated Conditional Modes (ICM) algorithm. Finally, experiments indicate that this algorithm for the cloud image detection has higher average accuracy rate which is about 98.76% and the average result can also reach 96.92% for different type of image noise.