Search

In 1979 Haralick famously introduced a method for analyzing the texture of an image: a set of statistics extracted from the co-occurrence matrix. In this paper we investigate novel sets of texture descriptors extracted from the co-occurrence matrix; in addition, we compare and combine different strategies for extending these descriptors. The following approaches are compared: the standard approach proposed by Haralick, two methods that consider the co-occurrence matrix as a three-dimensional shape, a gray-level run-length set of features and the direct use of the co-occurrence matrix projected onto a lower dimensional subspace by principal component analysis. Texture descriptors are extracted from the co-occurrence matrix evaluated at multiple scales. Moreover, the descriptors are extracted not only from the entire co-occurrence matrix but also from subwindows. The resulting texture descriptors are used to train a support vector machine and ensembles. Results show that our novel extraction methods improve the performance of standard methods. We validate our approach across six medical datasets representing different image classification problems using the Wilcoxon signed rank test. The source code used for the approaches tested in this paper will be available at: http://www.dei.unipd.it/wdyn/?IDsezione=3314&IDgruppo_pass=124&preview=. [ABSTRACT FROM AUTHOR]

In some large-scale face recognition task, such as driver license identification and law enforcement, the training set only contains one image per person. This situation is referred to as one sample problem. Because many face recognition techniques implicitly assume that several (at least two) images per person are available for training, they cannot deal with the one sample problem. This paper investigates principal component analysis (PCA), Fisher linear discriminant analysis (LDA), and locality preserving projections (LPP) and shows why they cannot perform well in one sample problem. After that, this paper presents four reasons that make one sample problem itself difficult: the small sample size problem; the lack of representative samples; the underestimated intra-class variation; and the overestimated inter-class variation. Based on the analysis, this paper proposes to enlarge the training set based on the inter-class relationship. This paper also extends LDA and LPP to extract features from the enlarged training set. The experimental results show the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The Student's-t mixture model, which is heavily tailed and more robust than the Gaussian mixture model, has recently received great attention on image processing. In this paper, we propose a robust non-rigid point set registration algorithm using the Student's-t mixture model. Specifically, first, we consider the alignment of two point sets as a probability density estimation problem and treat one point set as Student's-t mixture model centroids. Then, we fit the Student's-t mixture model centroids to the other point set which is treated as data. Finally, we get the closed-form solutions of registration parameters, leading to a computationally efficient registration algorithm. The proposed algorithm is especially effective for addressing the non-rigid point set registration problem when significant amounts of noise and outliers are present. Moreover, less registration parameters have to be set manually for our algorithm compared to the popular coherent points drift (CPD) algorithm. We have compared our algorithm with other state-of-the-art registration algorithms on both 2D and 3D data with noise and outliers, where our non-rigid registration algorithm showed accurate results and outperformed the other algorithms. [ABSTRACT FROM AUTHOR]

Using statistical textons for texture classification has shown great success recently. The maximal response 8 (Statistical_MR8), image patch (Statistical_Joint) and locally invariant fractal (Statistical_Fractal) are typical statistical texton algorithms and state-of-the-art texture classification methods. However, there are two limitations when using these methods. First, it needs a training stage to build a texton library, thus the recognition accuracy will be highly depended on the training samples; second, during feature extraction, local feature is assigned to a texton by searching for the nearest texton in the whole library, which is time consuming when the library size is big and the dimension of feature is high. To address the above two issues, in this paper, three binary texton counterpart methods were proposed, Binary_MR8, Binary_Joint, and Binary_Fractal. These methods do not require any training step but encode local feature into binary representation directly. The experimental results on the CUReT, UIUC and KTH-TIPS databases show that binary texton could get sound results with fast feature extraction, especially when the image size is not big and the quality of image is not poor. [ABSTRACT FROM AUTHOR]

Posture segmentation plays an essential role in human motion analysis. The state-of-the-art method extracts sufficiently high-dimensional features from 3D depth images for each 3D point and learns an efficient body part classifier. However, high-dimensional features are memory-consuming and difficult to handle on large-scale training dataset. In this paper, we propose an efficient two-stage dimension reduction scheme, termed biview learning, to encode two independent views which are depth-difference features (DDF) and relative position features (RPF). Biview learning explores the complementary property of DDF and RPF, and uses two stages to learn a compact yet comprehensive low-dimensional feature space for posture segmentation. In the first stage, discriminative locality alignment (DLA) is applied to the high-dimensional DDF to learn a discriminative low-dimensional representation. In the second stage, canonical correlation analysis (CCA) is used to explore the complementary property of RPF and the dimensionality reduced DDF. Finally, we train a support vector machine (SVM) over the output of CCA. We carefully validate the effectiveness of DLA and CCA utilized in the two-stage scheme on our 3D human points cloud dataset. Experimental results show that the proposed biview learning scheme significantly outperforms the state-of-the-art method for human posture segmentation. [ABSTRACT FROM AUTHOR]

Purpose: This paper describes a novel method to automatically segment the human brainstem into midbrain and pons, called LABS: Landmark-based Automated Brainstem Segmentation. LABS processes high-resolution structural magnetic resonance images (MRIs) according to a revised landmark-based approach integrated with a thresholding method, without manual interaction. Methods: This method was first tested on morphological T1-weighted MRIs of 30 healthy subjects. Its reliability was further confirmed by including neurological patients (with Alzheimer's Disease) from the ADNI repository, in whom the presence of volumetric loss within the brainstem had been previously described. Segmentation accuracies were evaluated against expert-drawn manual delineation. To evaluate the quality of LABS segmentation we used volumetric, spatial overlap and distance-based metrics. Results: The comparison between the quantitative measurements provided by LABS against manual segmentations revealed excellent results in healthy controls when considering either the midbrain (DICE measures higher that 0.9; Volume ratio around 1 and Hausdorff distance around 3) or the pons (DICE measures around 0.93; Volume ratio ranging 1.024–1.05 and Hausdorff distance around 2). Similar performances were detected for AD patients considering segmentation of the pons (DICE measures higher that 0.93; Volume ratio ranging from 0.97–0.98 and Hausdorff distance ranging 1.07–1.33), while LABS performed lower for the midbrain (DICE measures ranging 0.86–0.88; Volume ratio around 0.95 and Hausdorff distance ranging 1.71–2.15). Conclusions: Our study represents the first attempt to validate a new fully automated method for in vivo segmentation of two anatomically complex brainstem subregions. We retain that our method might represent a useful tool for future applications in clinical practice. [ABSTRACT FROM AUTHOR]

Image encryption is an important and effective technique to protect image security. In this paper, a novel image encryption algorithm combining Julia sets and Hilbert curves is proposed. The algorithm utilizes Julia sets’ parameters to generate a random sequence as the initial keys and gets the final encryption keys by scrambling the initial keys through the Hilbert curve. The final cipher image is obtained by modulo arithmetic and diffuse operation. In this method, it needs only a few parameters for the key generation, which greatly reduces the storage space. Moreover, because of the Julia sets’ properties, such as infiniteness and chaotic characteristics, the keys have high sensitivity even to a tiny perturbation. The experimental results indicate that the algorithm has large key space, good statistical property, high sensitivity for the keys, and effective resistance to the chosen-plaintext attack. [ABSTRACT FROM AUTHOR]

The part-based method has been a fast rising framework for object detection. It is attracting more and more attention for its detection precision and partial robustness to the occlusion. However, little research has been focused on the problem of occlusion overlapping of the part regions, which can reduce the performance of the system. This paper proposes a part-based probabilistic model and the corresponding inference algorithm for the problem of the part occlusion. The model is based on the Bayesian theory integrally and aims to be robust to the large occlusion. In the stage of the model construction, all of the parts constitute the vertex set of a fully connected graph, and a binary variable is assigned to each part to indicate its occlusion status. In addition, we introduce a penalty term to regularize the argument space of the objective function. Thus, the part detection is formulated as an optimization problem, which is divided into two alternative procedures: the outer inference and the inner inference. A stochastic tentative method is employed in the outer inference to determine the occlusion status for each part. In the inner inference, the gradient descent algorithm is employed to find the optimal positions of the parts, in term of the current occlusion status. Experiments were carried out on the Caltech database. The results demonstrated that the proposed method achieves a strong robustness to the occlusion. [ABSTRACT FROM AUTHOR]

Accurate and robust brain extraction is a critical step in most neuroimaging analysis pipelines. In particular, for the large-scale multi-site neuroimaging studies involving a significant number of subjects with diverse age and diagnostic groups, accurate and robust extraction of the brain automatically and consistently is highly desirable. In this paper, we introduce population-specific probability maps to guide the brain extraction of diverse subject groups, including both healthy and diseased adult human populations, both developing and aging human populations, as well as non-human primates. Specifically, the proposed method combines an atlas-based approach, for coarse skull-stripping, with a deformable-surface-based approach that is guided by local intensity information and population-specific prior information learned from a set of real brain images for more localized refinement. Comprehensive quantitative evaluations were performed on the diverse large-scale populations of ADNI dataset with over 800 subjects (55∼90 years of age, multi-site, various diagnosis groups), OASIS dataset with over 400 subjects (18∼96 years of age, wide age range, various diagnosis groups), and NIH pediatrics dataset with 150 subjects (5∼18 years of age, multi-site, wide age range as a complementary age group to the adult dataset). The results demonstrate that our method consistently yields the best overall results across almost the entire human life span, with only a single set of parameters. To demonstrate its capability to work on non-human primates, the proposed method is further evaluated using a rhesus macaque dataset with 20 subjects. Quantitative comparisons with popularly used state-of-the-art methods, including BET, Two-pass BET, BET-B, BSE, HWA, ROBEX and AFNI, demonstrate that the proposed method performs favorably with superior performance on all testing datasets, indicating its robustness and effectiveness. [ABSTRACT FROM AUTHOR]

Digital reconstruction of three-dimensional (3D) neuronal morphology from light microscopy images provides a powerful technique for analysis of neural circuits. It is time-consuming to manually perform this process. Thus, efficient computer-assisted approaches are preferable. In this paper, we present an innovative method for the tracing and reconstruction of 3D neuronal morphology from light microscopy images. The method uses a prediction and refinement strategy that is based on exploration of local neuron structural features. We extended the rayburst sampling algorithm to a marching fashion, which starts from a single or a few seed points and marches recursively forward along neurite branches to trace and reconstruct the whole tree-like structure. A local radius-related but size-independent hemispherical sampling was used to predict the neurite centerline and detect branches. Iterative rayburst sampling was performed in the orthogonal plane, to refine the centerline location and to estimate the local radius. We implemented the method in a cooperative 3D interactive visualization-assisted system named flNeuronTool. The source code in C++ and the binaries are freely available at http://sourceforge.net/projects/flneurontool/. We validated and evaluated the proposed method using synthetic data and real datasets from the Digital Reconstruction of Axonal and Dendritic Morphology (DIADEM) challenge. Then, flNeuronTool was applied to mouse brain images acquired with the Micro-Optical Sectioning Tomography (MOST) system, to reconstruct single neurons and local neural circuits. The results showed that the system achieves a reasonable balance between fast speed and acceptable accuracy, which is promising for interactive applications in neuronal image analysis. [ABSTRACT FROM AUTHOR]

Imaging processing techniques have been shown to be useful in studying protein domain structures. The idea is to represent the pairwise distances of any two residues of the structure in a 2D distance matrix (DM). Features and/or submatrices are extracted from this DM to represent a domain. Existing approaches, however, may involve a large number of features (100–400) or complicated mathematical operations. Finding fewer but more effective features is always desirable. In this paper, based on some key observations on DMs, we are able to decompose a DM image into four basic binary images, each representing the structural characteristics of a fundamental secondary structure element (SSE) or a motif in the domain. Using the concept of moments in image processing, we further derive 45 structural features based on the four binary images. Together with 4 features extracted from the basic images, we represent the structure of a domain using 49 features. We show that our feature vectors can represent domain structures effectively in terms of the following. (1) We show a higher accuracy for domain classification. (2) We show a clear and consistent distribution of domains using our proposed structural vector space. (3) We are able to cluster the domains according to our moment features and demonstrate a relationship between structural variation and functional diversity. [ABSTRACT FROM AUTHOR]

Minimum squared error based classification (MSEC) method establishes a unique classification model for all the test samples. However, this classification model may be not optimal for each test sample. This paper proposes an improved MSEC (IMSEC) method, which is tailored for each test sample. The proposed method first roughly identifies the possible classes of the test sample, and then establishes a minimum squared error (MSE) model based on the training samples from these possible classes of the test sample. We apply our method to face recognition. The experimental results on several datasets show that IMSEC outperforms MSEC and the other state-of-the-art methods in terms of accuracy. [ABSTRACT FROM AUTHOR]

Currently, remote sensing technologies were widely employed in the dynamic monitoring of the land. This paper presented an algorithm named fuzzy nonlinear proximal support vector machine (FNPSVM) by basing on ETM+ remote sensing image. This algorithm is applied to extract various types of lands of the city Da’an in northern China. Two multi-category strategies, namely “one-against-one” and “one-against-rest” for this algorithm were described in detail and then compared. A fuzzy membership function was presented to reduce the effects of noises or outliers on the data samples. The approaches of feature extraction, feature selection, and several key parameter settings were also given. Numerous experiments were carried out to evaluate its performances including various accuracies (overall accuracies and kappa coefficient), stability, training speed, and classification speed. The FNPSVM classifier was compared to the other three classifiers including the maximum likelihood classifier (MLC), back propagation neural network (BPN), and the proximal support vector machine (PSVM) under different training conditions. The impacts of the selection of training samples, testing samples and features on the four classifiers were also evaluated in these experiments. [ABSTRACT FROM AUTHOR]

This paper describes an automatic algorithm that uses a geometry-driven optimization approach to restore the shape of three-dimensional (3D) left ventricular (LV) models created from magnetic resonance imaging (MRI) data. The basic premise is to restore the LV shape such that the LV epicardial surface is smooth after the restoration and that the general shape characteristic of the LV is not altered. The Maximum Principle Curvature () and the Minimum Principle Curvature () of the LV epicardial surface are used to construct a shape-based optimization objective function to restore the shape of a motion-affected LV via a dual-resolution semi-rigid deformation process and a free-form geometric deformation process. A limited memory quasi-Newton algorithm, L-BFGS-B, is then used to solve the optimization problem. The goal of the optimization is to achieve a smooth epicardial shape by iterative in-plane and through-plane translation of vertices in the LV model. We tested our algorithm on 30 sets of LV models with simulated motion artifact generated from a very smooth patient sample, and 20 in vivo patient-specific models which contain significant motion artifacts. In the 30 simulated samples, the Hausdorff distances with respect to the Ground Truth are significantly reduced after restoration, signifying that the algorithm can restore geometrical accuracy of motion-affected LV models. In the 20 in vivo patient-specific models, the results show that our method is able to restore the shape of LV models without altering the general shape of the model. The magnitudes of in-plane translations are also consistent with existing registration techniques and experimental findings. [ABSTRACT FROM AUTHOR]

Tensor subspace transformation, a commonly used subspace transformation technique, has gained more and more popularity over the past few years because many objects in the real world can be naturally represented as multidimensional arrays, i.e. tensors. For example, a RGB facial image can be represented as a three-dimensional array (or 3rd-order tensor). The first two dimensionalities (or modes) represent the facial spatial information and the third dimensionality (or mode) represents the color space information. Each mode of the tensor may express a different semantic meaning. Thus different transformation strategies should be applied to different modes of the tensor according to their semantic meanings to obtain the best performance. To the best of our knowledge, there are no existing tensor subspace transformation algorithm which implements different transformation strategies on different modes of a tensor accordingly. In this paper, we propose a fusion tensor subspace transformation framework, a novel idea where different transformation strategies are implemented on separate modes of a tensor. Under the framework, we propose the Fusion Tensor Color Space (FTCS) model for face recognition. [ABSTRACT FROM AUTHOR]

In this paper, a Bregman iteration based total variation image restoration algorithm is proposed. Based on the Bregman iteration, the algorithm splits the original total variation problem into sub-problems that are easy to solve. Moreover, non-local regularization is introduced into the proposed algorithm, and a method to choose the non-local filter parameter locally and adaptively is proposed. Experiment results show that the proposed algorithms outperform some other regularization methods. [ABSTRACT FROM AUTHOR]

This paper examines the multiple atlas random diffeomorphic orbit model in Computational Anatomy (CA) for parameter estimation and segmentation of subcortical and ventricular neuroanatomy in magnetic resonance imagery. We assume that there exist multiple magnetic resonance image (MRI) atlases, each atlas containing a collection of locally-defined charts in the brain generated via manual delineation of the structures of interest. We focus on maximum a posteriori estimation of high dimensional segmentations of MR within the class of generative models representing the observed MRI as a conditionally Gaussian random field, conditioned on the atlas charts and the diffeomorphic change of coordinates of each chart that generates it. The charts and their diffeomorphic correspondences are unknown and viewed as latent or hidden variables. We demonstrate that the expectation-maximization (EM) algorithm arises naturally, yielding the likelihood-fusion equation which the a posteriori estimator of the segmentation labels maximizes. The likelihoods being fused are modeled as conditionally Gaussian random fields with mean fields a function of each atlas chart under its diffeomorphic change of coordinates onto the target. The conditional-mean in the EM algorithm specifies the convex weights with which the chart-specific likelihoods are fused. The multiple atlases with the associated convex weights imply that the posterior distribution is a multi-modal representation of the measured MRI. Segmentation results for subcortical and ventricular structures of subjects, within populations of demented subjects, are demonstrated, including the use of multiple atlases across multiple diseased groups. [ABSTRACT FROM AUTHOR]

Most digital cameras use an array of alternating color filters to capture the varied colors in a scene with a single sensor chip. Reconstruction of a full color image from such a color mosaic is what constitutes demosaicing. In this paper, a technique is proposed that performs this demosaicing in a way that incurs a very low computational cost. This is done through a (dual-tree complex) wavelet interpretation of the demosaicing problem. By using a novel locally adaptive approach for demosaicing (complex) wavelet coefficients, we show that many of the common demosaicing artifacts can be avoided in an efficient way. Results demonstrate that the proposed method is competitive with respect to the current state of the art, but incurs a lower computational cost. The wavelet approach also allows for computationally effective denoising or deblurring approaches. [ABSTRACT FROM AUTHOR]

The one-sample-per-person problem has become an active research topic for face recognition in recent years because of its challenges and significance for real-world applications. However, achieving relatively higher recognition accuracy is still a difficult problem due to, usually, too few training samples being available and variations of illumination and expression. To alleviate the negative effects caused by these unfavorable factors, in this paper we propose a more accurate spectral feature image-based 2DLDA (two-dimensional linear discriminant analysis) ensemble algorithm for face recognition, with one sample image per person. In our algorithm, multi-resolution spectral feature images are constructed to represent the face images; this can greatly enlarge the training set. The proposed method is inspired by our finding that, among these spectral feature images, features extracted from some orientations and scales using 2DLDA are not sensitive to variations of illumination and expression. In order to maintain the positive characteristics of these filters and to make correct category assignments, the strategy of classifier committee learning (CCL) is designed to combine the results obtained from different spectral feature images. Using the above strategies, the negative effects caused by those unfavorable factors can be alleviated efficiently in face recognition. Experimental results on the standard databases demonstrate the feasibility and efficiency of the proposed method. [ABSTRACT FROM AUTHOR]