Search

In the last decades, Gaussian Mixture Models (GMMs) have attracted considerable interest in data mining and pattern recognition. A GMM-based clustering algorithm models a dataset with a mixture of multiple Gaussian components and estimates the model parameters using the Expectation-Maximization (EM) algorithm. Recently, a new Locally Consistent GMM (LCGMM) has been proposed to improve the clustering performance by exploiting the local manifold structure of the data using a p nearest neighbor graph. In addition to the underlying manifold structure, many other forms of prior knowledge may guide the clustering process and improve the performance. In this paper, we introduce a Semi-Supervised LCGMM (Semi-LCGMM), where the prior knowledge is provided in the form of class labels of partial data. In particular, the new Semi-LCGMM incorporates the prior knowledge into the maximum likelihood function of the original LCGMM, and the model parameters are estimated using the EM algorithm. It is worth noting that, in our algorithm, each class may be modeled by multiple Gaussian components while in the unsupervised setting each class is modeled by a single Gaussian component. Our algorithm has shown promising results in many different applications, including clustering breast cancer data, heart disease data, handwritten digit images, human face images, and image segmentation.

This paper presents a novel object detection method using a single instance from the object category. Our method uses biologically inspired global scene context criteria to check whether every individual location of the image can be naturally replaced by the query instance, which indicates whether there is a similar object at this location. Different from the traditional detection methods that only look at individual locations for the desired objects, our method evaluates the consistency of the entire scene. It is therefore robust to large intra-class variations, occlusions, a minor variety of poses, low-revolution conditions, background clutter etc., and there is no off-line training. The experimental results on four datasets and two video sequences clearly show the superior robustness of the proposed method, suggesting that global scene context is important for visual detection/localization.

Face recognition is one of the most important applications of machine learning and computer vision. The traditional supervised learning methods require a large amount of labeled face images to achieve good performance. In practice, however, labeled images are usually scarce while unlabeled ones may be abundant. In this paper, we introduce a semi-supervised face recognition method, in which semi-supervised linear discriminant analysis (SDA) and affinity propagation (AP) are integrated into a self-training framework. In particular, SDA is employed to compute the face subspace using both labeled and unlabeled images, and AP is used to identify the exemplars of different face classes in the subspace. The unlabeled data can then be classified according to the exemplars and the newly labeled data with the highest confidence are added to the labeled data, and the whole procedure iterates until convergence. A series of experiments on four face datasets are carried out to evaluate the performance of our algorithm. Experimental results illustrate that our algorithm outperforms the other unsupervised, semi-supervised, and supervised methods.

It is essential to extract dynamic information about a patient's heart from a medical X-ray angiography image sequence for a quantitative medical diagnosis. As the motions included in the angiography sequences are the mixture of various motion signals, such as the body's integral translation, respiratory motion, cardiac impulse, and tremor, automatic separation of these signals challenges the effectiveness of the image information processing method. This paper has proposed an optimal time-frequency domains iteration separation algorithm for multi-motion parameters (TFISA-MMP) to obtain the physiological parameters including the translation. The main procedures of the TFISA-MMP algorithm include three parts. First, the algorithm automatically extracted a set of relatively stable branch points from the coronary artery angiography image, and then automatically tracked these branch points in the sequence image to obtain their motion curves that changed with the time. Second, with the guidance of the multi-motion parameter model, the initial values of each component were estimated based on Discrete Fourier Transformation (DFT). Moreover, the initial values of each motion component were optimized using the global mean square minimum error between the estimated reconstructed signal and original signal and the local mean square minimum error in the frequency domain for each frequency component. Finally, these motion components were estimated by minimizing the residual signal between the original signal and the reconstructed signal via the loop iteration to obtain the estimated optimal motion components, such as two-dimensional (2D) heartbeat, tremor, respiratory motion, and translational motion. Both visible human coronary model simulation experiments and the clinical experiments of single-arm X-ray angiography images of many individuals verified the correctness, validity, and clinical applicability of the separation method. [ABSTRACT FROM AUTHOR]

The goal of this work is to present a computational model for contour detection, based on the surround suppression mechanisms of the primary visual cortex, in which the strength of surround suppression can adaptively vary with contrast—the surround modulation tends to be clearly suppressive at high contrast and less suppressive at low contrast, which may help to achieve the tradeoff between reducing cluttered edges and retaining object structures with weak responses, thus improving performance of contour detection under different circumstances; on the other hand, for similar texels can engender stronger suppression effects, more than orientation feature, texton is introduced to homogeneity measurement, which can well character multiple properties of texture regions, thereby yielding better contour detection results than other suppression models. The study can provide useful suggestions for contour detection algorithm in computer vision, but may also contribute to understanding contrast influence and feature representation in non-classical receptive field inhibition. [ABSTRACT FROM AUTHOR]

Abstract: An adaptive enhancement algorithm is proposed in the paper for low illumination color image which is low brightness and low contrast. The algorithm is realized through three steps: the global brightness adaptive adjustment, locally adaptive contrast enhancement and color restore. Experiments show that the brightness and local contrast of low illumination color image can be effectively enhanced. The details, especially in the dark region of enhanced image are more prominent and the enhanced image is more vivid than some other methods for low illumination color image enhancement. [Copyright &y& Elsevier]

Abstract: We propose a new method to estimate image bias field by an intuitive assumption that the overall trend of image surface is associated with bias field and the true bias field can evolve from image surface. To constrain the evolution, an energy function is designed based on two diffusion constraints. One is that the bias-free image has concise representation with image gradients. The other constraint is that model fit on the bias field which is more relaxed than previous methods. The new algorithm enables adaptive specifications of the estimated bias field's smoothness. The experimental results show our new technique can be applied to intensity inhomogeneity correction of medical image effectively. [Copyright &y& Elsevier]

The local binary pattern (LBP) approach has been widely used in texture description. In this paper, we build a new framework to extract the binary patterns and propose a robust texture descriptor: multi-ring local binary pattern (MrLBP). The MrLBP algorithm creates patterns from several ringed areas and mainly contains two parts. One is the extra-ring local binary pattern operator that gets patterns from the mean values of different ringed areas. The other is the intra-ring local binary pattern operator that obtains patterns by counting the majority of binary values in every single ringed area. Moreover, the binary formation of each part of the MrLBP is obtained from two different aspects. The MrLBP method not only considers the binary relationship among pixels in a local region, but also focuses on the relationship between pixels in a local region and the whole image. This is a little different from the conventional LBP methods that only get the binary formation from the local gray scales differences. The experimental results on two public databases have validated the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: Semi-supervised classification has become an active topic recently and a number of algorithms, such as Self-training, have been proposed to improve the performance of supervised classification using unlabeled data. In this paper, we propose a semi-supervised learning framework which combines clustering and classification. Our motivation is that clustering analysis is a powerful knowledge-discovery tool and it may reveal the underlying data space structure from unlabeled data. In our framework, semi-supervised clustering is integrated into Self-training classification to help train a better classifier. In particular, the semi-supervised fuzzy c-means algorithm and support vector machines are used for clustering and classification, respectively. Experimental results on artificial and real datasets demonstrate the advantages of the proposed framework. [Copyright &y& Elsevier]

Abstract: Automatic image segmentation is always a fundamental but challenging problem in computer vision. The simplest approach to image segmentation may be clustering feature vectors of pixels at first, then labeling each pixel with its corresponding cluster. This requires that the clustering on feature space must be robust. However, most of popular clustering algorithms could not obtain a robust clustering result yet, if the clusters in feature space have a complex distribution. Generally, for most of clustering-based segmentation methods, it still needs more constraints of positional relations between pixels in image lattice to be utilized during the procedure of clustering. Our works in this paper address the problem of image segmentation under the paradigm of pure clustering-then-labeling. A robust clustering algorithm which could maintain good coherence of data in feature space is proposed and utilized to do clustering on the L∗a∗b∗ color feature space of pixels. Image segmentation is straightforwardly obtained by setting each pixel with its corresponding cluster. Further, based on the theory of Minimum Description Length, an effective approach to automatic parameter selection for our segmentation method is also proposed. We test our segmentation method on Berkeley segmentation database, and the experimental results show that our method compares favorably against some state-of-the-art segmentation methods. [Copyright &y& Elsevier]

Abstract: Despite of its great success, two key problems are still unresolved for AdaBoost algorithms: how to select the most discriminative weak learners and how to optimally combine them. In this paper, a new AdaBoost algorithm is proposed to make improvement in the two aspects. First, we select the most discriminative weak learners by minimizing a novel distance related criterion, i.e., error-degree-weighted training error metric (ETEM) together with generalization capability metric (GCM), rather than training error rate only. Second, after getting the coefficients that are set empirically, we combine the weak learners optimally by tuning the coefficients using kernel-based perceptron. Experiments with synthetic and real scene data sets show our algorithm outperforms conventional AdaBoost. [Copyright &y& Elsevier]

In this paper, we propose a new method for online multiple people tracking, which combines the detection process and the single object tracking process, and establishes the interactions between them. The detector detects objects in the still images which ignores the sequential information. Meantime, the single object tracker does not use the category semantic information during tracking. To take both the sequential and semantic information into account, we exchange information among the detector and the trackers. More specifically, the trackers deliver sequential information to the detector by providing the detector with the extra proposals. The detector supplements each tracker with the robust semantic information by using bounding box regression to modify the tracking result. Besides, the interactions also happen among the trackers through the occlusion speculation, the perspective model interpretation and the trajectory merging process. The experimental results demonstrate that the proposed algorithm performs favorably against the state-of-the-art MOT methods. [ABSTRACT FROM AUTHOR]

Abstract: Factor analysis is a powerful tool used for the analysis of dynamic studies. One of the major drawbacks of Factor Analysis of Dynamic Structures (FADS) is that the solution is not mathematically unique when only non-negativity constraints are used to determine factors and factor coefficients. In this paper, we introduce a novel method to correct FADS solutions by constructing and minimizing a new objective function. The method is improved from non-negative matrix factorizations (NMFs) algorithm by adding a sparse constraint that penalizes multiple components in the images of the factor coefficients. The technique is tested on computer simulations, and a patient ultrasound liver study. The results show that the method works well in comparison to the truth in computer simulations and to region of interest (ROI) measurements in the experimental studies. [Copyright &y& Elsevier]

Abstract: As an optical image encryption technique, the double random phase encoding (DRPE) technique can be used to encrypt the watermark image. However, since the DRPE system may be attacked with cryptanlysis, the security of the DRPE technique is somewhat weak. To improve the security, an image watermark generation algorithm based on cascaded iterative Fourier transform algorithm (CIFT) was proposed. In the proposed algorithm, two phase masks were obtained by applying the CIFT algorithm on the original reference watermark image and the final embedded watermark image was obtained with them. One of the phase masks was kept as the secret key. When detecting the watermark, the embedded watermark was extracted from the watermarked host image. Then, the recovered watermarked was obtained with the extracted watermark and the secret key. By comparing the recovered watermark and the original reference watermark, whether the tested image containing the watermark or not can be judged. With the CIFT algorithm, it is more secure to generate the embedded watermark. In addition, it is practical because only one of the phase masks is needed for watermark recovery. The simulation results show that the proposed method has good imperceptibility and robustness. [Copyright &y& Elsevier]