Search

Many methods have been developed to improve the performance of image registration. In this letter, we introduce a novel method based on a local affine constraint for remote sensing image registration, which can be widely used in image processing and pattern recognition. Our algorithm has three components. First, we exploit the scale invariant feature transform (SIFT) method to extract feature points and calculate the gradient magnitude to establish feature descriptors with a circular instead of square neighborhood. Second, an initial matching is implemented by the nearest neighbor distance ratio (NNDR) and the fast sample consensus (FSC) algorithm. Finally, fine registration is established using more correct matches obtained by the local affine transformation circular region search algorithm. Experimental results show that the proposed method achieves subpixel accuracy. In addition, both the correct matching rate and registration demonstrate the effectiveness and efficiency of our method.

Being able to collect rich morphological information of brain, structural magnetic resonance imaging (MRI) is popularly applied to computer-aided diagnosis of Alzheimer’s disease (AD). Conventional methods for AD diagnosis are labor-intensive and typically depend on a substantial amount of hand-crafted features. In this paper, we propose a novel framework of convolutional neural network that aims at identifying AD or normal control, and mild cognitive impairment or normal control. The centerpiece of our method are pseudo-3D block and expanded global context block which are integrated into residual block of backbone in a cascaded manner. To be specific, we transfer pseudo-3D block in the video feature representation to extract structural MRI features. Besides, we extend the 2D global context block to the 3D model which can effectively combine the features and capture the latent associations, while simulate the global context in each dimension of structural MRI results. With the preprocessed structural MRI used as the input of the overall network, our method is capable of constructing a latent representation with multiple residual blocks to promote the classification accuracy. Intrinsically, our method reduces the complexity of conventional 3D convolutional network model applied to AD diagnosis and improves the classification accuracy over the baseline. Furthermore, our network can fully take advantage of the deep 3D convolutional neural network for automatic feature extraction and representation, and thus avoids the limitation of low processing efficiency caused by the preprocessing procedure in which a specific area needs to be annotated in advance. Experimental results on Alzheimer’s Disease Neuroimaging Initiative database indicate that our proposed method reports accuracy of 89.29% on the AD/NC and 87.57% on the mild cognitive impairment/NC, whilst our approach achieves the 0.5% improvement of accuracy compared with the backbone.

Considering a wide range of applications of nonnegative matrix factorization (NMF), many NMF and their variants have been developed. Since previous NMF methods cannot fully describe complex inner global and local manifold structures of the data space and extract complex structural information, we propose a novel NMF method called dual-graph global and local concept factorization (DGLCF). To properly describe the inner manifold structure, DGLCF introduces the global and local structures of the data manifold and the geometric structure of the feature manifold into CF. The global manifold structure makes the model more discriminative, while the two local regularization terms simultaneously preserve the inherent geometry of data and features. Finally, we analyze convergence and the iterative update rules of DGLCF. We illustrate clustering performance by comparing it with latest algorithms on four real-world datasets.

This paper presents a novel medical image registration algorithm named total variation constrained graph-regularization for non-negative matrix factorization (TV-GNMF). The method utilizes non-negative matrix factorization by total variation constraint and graph regularization. The main contributions of our work are the following. First, total variation is incorporated into NMF to control the diffusion speed. The purpose is to denoise in smooth regions and preserve features or details of the data in edge regions by using a diffusion coefficient based on gradient information. Second, we add graph regularization into NMF to reveal intrinsic geometry and structure information of features to enhance the discrimination power. Third, the multiplicative update rules and proof of convergence of the TV-GNMF algorithm are given. Experiments conducted on datasets show that the proposed TV-GNMF method outperforms other state-of-the-art algorithms.

We propose a novel NMF algorithm, named Total Variation constrained Graph-regularized Convex Non-negative Matrix Factorization (TV-GCNMF), to incorporate total variation and graph Laplacian with convex NMF. In this model, the feature details of the data are preserved by a diffusion coefficient based on the gradient information. The graph regularization and convex constraints reveal the intrinsic geometry and structure information of the features; thereby, obtaining sparse and parts-based representations. Furthermore, we give the multiplicative update rules and prove convergence of the proposed algorithm. The results of clustering experiments on multiple datasets, under various noise conditions, show the effectiveness and robustness of the proposed method compared to state-of-the-art clustering methods and other related work.

Image registration has always been an important research topic. This paper proposes a novel method of constructing descriptors called the histogram of oriented local binary pattern descriptor (HOLBP) for fast and robust matching. There are three new components in our algorithm. First, we redefined the gradient and angle calculation template to make it more sensitive to edge information. Second, we proposed a new construction method of the HOLBP descriptor and improved the traditional local binary pattern (LBP) computation template. Third, the principle of uniform rotation-invariant LBP was applied to add 10-dimensional gradient direction information to form a 138-dimension HOLBP descriptor vector. The experimental results showed that our method is very stable in terms of accuracy and computational time for different test images.

This paper proposes a Graph regularized Lp smooth non-negative matrix factorization &#x0028 GSNMF &#x0029 method by incorporating graph regularization and Lp smoothing constraint, which considers the intrinsic geometric information of a data set and produces smooth and stable solutions. The main contributions are as follows: first, graph regularization is added into NMF to discover the hidden semantics and simultaneously respect the intrinsic geometric structure information of a data set. Second, the Lp smoothing constraint is incorporated into NMF to combine the merits of isotropic &#x0028 L2-norm &#x0029 and anisotropic &#x0028 L1-norm &#x0029 diffusion smoothing, and produces a smooth and more accurate solution to the optimization problem. Finally, the update rules and proof of convergence of GSNMF are given. Experiments on several data sets show that the proposed method outperforms related state-of-the-art methods.

Retracted.

Image registration is an important technique in many computer vision applications such as image fusion, image retrieval, object tracking, face recognition, change detection and so on. Local feature descriptors, i.e., how to detect features and how to describe them, play a fundamental and important role in image registration process, which directly influence the accuracy and robustness of image registration. This paper mainly focuses on the variety of local feature descriptors including some theoretical research, mathematical models, and methods or algorithms along with their applications in the context of image registration. The existing local feature descriptors are roughly classified into six categories to demonstrate and analyze comprehensively their own advantages. The current and future challenges of local feature descriptors are discussed. The major goal of the paper is to present a unique survey of the state-of-the-art image matching methods based on feature descriptor, from which future research may benefit.

Non-negative matrix factorization (NMF) has been widely applied in information retrieval and computer vision. However, its performance has been restricted due to its limited tolerance to data noise, as well as its inflexibility in setting regularization parameters. In this paper, we propose a novel sparse matrix factorization method for data representation to solve these problems, termed Adaptive Total-Variation Constrained based Non-Negative Matrix Factorization on Manifold (ATV-NMF). The proposed ATV can adaptively choose the anisotropic smoothing scheme based on the gradient information of data to denoise or preserve feature details by incorporating adaptive total variation into the factorization process. Notably, the manifold graph regularization is also incorporated into NMF, which can discover intrinsic geometrical structure of data to enhance the discriminability. Experimental results demonstrate that the proposed method is very effective for data clustering in comparison to the state-of-the-art algorithms on several standard benchmarks.

In this paper, we present a locality-constrained nonnegative robust shape interaction (LNRSI) subspace clustering method. LNRSI integrates the local manifold structure of data into the robust shape interaction (RSI) in a unified formulation, which guarantees the locality and the low-rank property of the optimal affinity graph. Compared with traditional low-rank representation (LRR) learning method, LNRSI can not only pursuit the global structure of data space by low-rank regularization, but also keep the locality manifold, which leads to a sparse and low-rank affinity graph. Due to the clear block-diagonal effect of the affinity graph, LNRSI is robust to noise and occlusions, and achieves a higher rate of correct clustering. The theoretical analysis of the clustering effect is also discussed. An efficient solution based on linearized alternating direction method with adaptive penalty (LADMAP) is built for our method. Finally, we evaluate the performance of LNRSI on both synthetic data and real computer vision tasks, i.e., motion segmentation and handwritten digit clustering. The experimental results show that our LNRSI outperforms several state-of-the-art algorithms.

The social groups often represent the groups within which are dense connections and between which are sparse connections. The social groups are actually the clusters. In this paper, we propose a simple but powerful method to combine the content and link information of the social network in the class, and analyze the results of different graph clustering algorithms, our experimental result shows that social groups detection by Affinity Propagation algorithm outperforms than the other clustering algorithms. In addition, we analyze the centrality of the detected communities and find the groups detected can significantly help to improve the quality of the teaching and learning.

Image registration is a key problem in a variety of applications, such as computer vision, medical image processing, pattern recognition, etc., while the application of registration is limited by time consumption and the accuracy in the case of large pose differences. Aimed at these two kinds of problems, we propose a fast rotation-free feature-based rigid registration method based on our proposed accelerated-NSIFT and GMM registration-based parallel optimization (PO-GMMREG). Our method is accelerated by using the GPU/CUDA programming and preserving only the location information without constructing the descriptor of each interest point, while its robustness to missing correspondences and outliers is improved by converting the interest point matching to Gaussian mixture model alignment. The accuracy in the case of large pose differences is settled by our proposed PO-GMMREG algorithm by constructing a set of initial transformations. Experimental results demonstrate that our proposed algorithm can fast rigidly register 3-D medical images and is reliable for aligning 3-D scans even when they exhibit a poor initialization.

Bioluminescence tomography (BLT) is a promising in vivo optical imaging technique in preclinical research at cellular and molecular levels. The problem of BLT reconstruction is quite ill-posed and ill-conditioned. In order to achieve high accuracy and efficiency for its inverse reconstruction, we proposed a novel approach based on L p regularization with the Split Bregman method. The diffusion equation was used as the forward model. Then, we defined the objective function of L p regularization and developed a Split Bregman iteration algorithm to optimize this function. After that, we conducted numerical simulations and in vivo experiments to evaluate the accuracy and efficiency of the proposed method. The results of the simulations indicated that compared with the conjugate gradient and iterative shrinkage methods, the proposed method is more accurate and faster for multisource reconstructions. Furthermore, in vivo imaging suggested that it could clearly distinguish the viable and apoptotic tumor regions. The Split Bregman iteration method is able to minimize the L p regularization problem and achieve fast and accurate reconstruction in BLT.

This paper presents a novel perspective on characterizing the spectral correspondence between the nodes of weighted graphs for image matching applications. The algorithm is based on the principal feature components obtained by stochastic perturbation of a graph. There are three areas of contributions in this paper. First, a stochastic normalized Laplacian matrix of a weighted graph is obtained by perturbing the matrix of a sensed graph model. Second, we obtain the eigenvectors based on an eigen-decomposition approach, where representative elements of each row of this matrix can be considered to be the feature components of a feature point. Third, correct correspondences are determined in a low-dimensional principal feature component space between the graphs. In order to further enhance image matching, we also exploit the random sample consensus algorithm, as a post-processing step, to eliminate mismatches in feature correspondences. The experiments on synthetic and real-world images demonstrate the effectiveness and accuracy of the proposed method.

We propose an incremental spectral clustering method for stream data clustering and apply it to stream image segmentation. The main idea in our work consists of generating the data points in the kernel space by Fastfood features and iteratively calculating the eigendecomposition of data. Compared with the popular Nystr&ouml, m-based approximation, our work accesses each data point only once while Nystr&ouml, m, in particular the sampling scheme, will go through the entire dataset first and calculate the embeddings of data points with a second visit. As a result, our method is able to learn data partitions incrementally and improve eigenvector approximation with more and more data seen from a stream. By contrast, the performance of the standard Nystr&ouml, m is fixed when the sample set is selected. Experimental results show the superiority of our method.

Estimating the transformation between two images from the same scene is a fundamental step for image registration, image stitching and 3D reconstruction. State-of-the-art methods are mainly based on sorted residual for generating hypotheses. This scheme has acquired encouraging results in many remote sensing applications. Unfortunately, mainstream residual based methods may fail in estimating the transform between Unmanned Aerial Vehicle (UAV) low altitude remote sensing images, due to the fact that UAV images always have repetitive patterns and severe viewpoint changes, which produce lower inlier rate and higher pseudo outlier rate than other tasks. We performed extensive experiments and found the main reason is that these methods compute feature pair similarity within a fixed window, making them sensitive to the size of residual window. To solve this problem, three schemes that based on the distribution of residuals are proposed, which are called Relational Window (RW), Sliding Window (SW), Reverse Residual Order (RRO), respectively. Specially, RW employs a relaxation residual window size to evaluate the highest similarity within a relaxation model length. SW fixes the number of overlap models while varying the length of window size. RRO takes the permutation of residual values into consideration to measure similarity, not only including the number of overlap structures, but also giving penalty to reverse number within the overlap structures. Experimental results conducted on our own built UAV high resolution remote sensing images show that the proposed three strategies all outperform traditional methods in the presence of severe perspective distortion due to viewpoint change.

Learning students social affinity and modeling their social networks are beneficial for instructors to design proper pedagogical strategies. Students seating distribution contains social data and can be used for analysing their social relationships. In this paper, we propose a method to automatically construct the class social network and predict the position of a student’s seat in class. First, we determine the positions of each student in a classroom by utilizing the center projection principle and linear fitting algorithms. The intimate relationship between students is captured to model their social network based on Euclidean distance. Then, we learn the social affinities from the Social Affinity Map (SAM) which clusters the relative positions of surrounding students. Based on this, students’ seating distribution can be predicted successfully with accuracy reaching 82.1%.

Non-negative matrix factorization (NMF) is a very popular dimensionality reduction method that has been widely used in computer vision and data clustering. However, NMF does not consider the intrinsic geometric information of a data set and also does not produce smooth and stable solutions. To resolve these problems, we propose a Graph regularized Lp Smooth Non-negative Matrix Factorization (GSNMF) method by incorporating graph regularization with Lp smooth constraint. The graph regularization can discover the hidden semantics and simultaneously respect the intrinsic geometric structure information of a data set. The Lp smooth constraint can combine the merits of isotropic (L2-norm) and anisotropic (L1-norm) diffusion smoothing, and produce a smooth and more accurate solution to the optimization problem. Experimental results on some data sets demonstrate that the proposed method outperforms related state-of-the-art NMF methods.