To prevent blindness from diabetic retinopathy, periodic screening and early diagnosis are neccessary. Due to lack of expert ophthalmologists in rural area, automated early exudate (one of visible sign of diabetic retinopathy) detection could help to reduce the number of blindness in diabetic patients. Traditional automatic exudate detection methods are based on specific parameter configuration, while the machine learning approaches which seems more flexible may be computationally high cost. A comparative analysis of traditional and machine learning of exudates detection, namely, mathematical morphology, fuzzy c-means clustering, naive Bayesian classifier, Support Vector Machine and Nearest Neighbor classifier are presented. Detected exudates are validated with expert ophthalmologists' hand-drawn ground-truths. The sensitivity, specificity, precision, accuracy and time complexity of each method are also compared.

In this study, we introduce shift-invariant sparse image representations using tree-structured dictionaries. Sparse coding is a generative signal model that approximates signals by the linear combinations of atoms in a dictionary. Since a sparsity penalty is introduced during signal approximation and dictionary learning, the dictionary represents the primal structures of the signals. Under the shift-invariance constraint, the dictionary comprises translated structuring elements (SEs). The computational cost and number of atoms in the dictionary increase along with the increasing number of SEs. In this paper, we propose an algorithm for shift-invariant sparse image representation, in which SEs are learnt with a tree-structured approach. By using a tree-structured dictionary, we can reduce the computational cost of the image decomposition to the logarithmic order of the number of SEs. We also present the results of our experiments on the SE learning and the use of our algorithm in image recovery applications.

In this paper, an image restoration method using the Wiener filter is proposed. In order to bring the theory of the Wiener filter consistent with images that have spatially varying statistics, the proposed method adopts the locally adaptive Wiener filter (AWF) based on the universal Gaussian mixture distribution model (UNI-GMM) previously proposed for denoising. Applying the UNI-GMM-AWF for deconvolution problem, the proposed method employs the stationary Wiener filter (SWF) as a pre-filter. The SWF in the discrete cosine transform domain shrinks the blur point spread function and facilitates the modeling and filtering at the proceeding AWF. The SWF and UNI-GMM are learned using a generic training image set and the proposed method is tuned toward the image set. Simulation results are presented to demonstrate the effectiveness of the proposed method.

This paper describes a novel algorithm for approximate nearest neighbor searching. For solving this problem especially in high dimensional spaces, one of the best-known algorithm is Locality-Sensitive Hashing (LSH). This paper presents a variant of the LSH algorithm that outperforms previously proposed methods when the dataset consists of vectors normalized to unit length, which is often the case in pattern recognition. The LSH scheme is based on a family of hash functions that preserves the locality of points. This paper points out that for our special case problem we can design efficient hash functions that map a point on the hypersphere into the closest vertex of the randomly rotated regular polytope. The computational analysis confirmed that the proposed method could improve the exponent ρ, the main indicator of the performance of the LSH algorithm. The practical experiments also supported the efficiency of our algorithm both in time and in space.

In this paper, an ER (Error-Reduction) algorithm-based method for removal of adherent water drops from images obtained by a rear view camera mounted on a vehicle in rainy conditions is proposed. Since Fourier-domain and object-domain constraints are needed for any ER algorithm-based method, the proposed method introduces the following two novel constraints for the removal of adherent water drops. The first one is the Fourier-domain constraint that utilizes the Fourier transform magnitude of the previous frame in the obtained images as that of the target frame. Noting that images obtained by the rear view camera have the unique characteristics of objects moving like ripples because the rear view camera is generally composed of a fish-eye lens for a wide view angle, the proposed method assumes that the Fourier transform magnitudes of the target frame and the previous frame are the same in the polar coordinate system. The second constraint is the object-domain constraint that utilizes intensities in an area of the target frame to which water drops have adhered. Specifically, the proposed method models a deterioration process of intensities that are corrupted by the water drop adhering to the rear view camera lens. By utilizing these novel constraints, the proposed ER algorithm can remove adherent water drops from images obtained by the rear view camera. Experimental results that verify the performance of the proposed method are represented.

Let G(s)=C(sI - A)-1B+D be a given system where entries of A,B,C,D are polynomials in a parameter k. Then H∞ norm || G(s) ||∞ of G(s) is a function of k, and [9] presents an algorithm to express 1/(||G(s) ||∞)2 as a root of a bivariate polynomial, assuming feedthrough term D to be zero. This paper extends the algorithm in two ways: The first extension is the form of the function to be expressed. The extended algorithm can treat, not only H∞ norm, but also functions that appear in the celebrated KYP Lemma. The other extension is the range of the frequency. While H∞ norm considers the supremum of the maximum singular value of G(i ω) for the infinite range 0 ≤ω ≤ ∞ of ω, the extended algorithm treats the norm for the finite frequency range ω ≤ ω ≤ ω- (ω, ω- ∈ R ∪ ∞). Those two extensions allow the algorithm to be applied to wider area of control problems. We give illustrative numerical examples where we apply the extended algorithm to the computation of the frequency-restricted norm, i.e., the supremum of the maximum singular value of G(i ω) (ω- ≤ ω ≤ ω-).

In this paper, a method for adaptive reconstruction of missing textures based on kernel canonical correlation analysis (CCA) with a new clustering scheme is presented. The proposed method estimates the correlation between two areas, which respectively correspond to a missing area and its neighboring area, from known parts within the target image and realizes reconstruction of the missing texture. In order to obtain this correlation, the kernel CCA is applied to each cluster containing the same kind of textures, and the optimal result is selected for the target missing area. Specifically, a new approach monitoring errors caused in the above kernel CCA-based reconstruction process enables selection of the optimal result. This approach provides a solution to the problem in traditional methods of not being able to perform adaptive reconstruction of the target textures due to missing intensities. Consequently, all of the missing textures are successfully estimated by the optimal cluster's correlation, which provides accurate reconstruction of the same kinds of textures. In addition, the proposed method can obtain the correlation more accurately than our previous works, and more successful reconstruction performance can be expected. Experimental results show impressive improvement of the proposed reconstruction technique over previously reported reconstruction techniques.

A blur restoration scheme for images with linear motion blurred objects in still background is proposed. The proposed scheme starts from a rough detection of locations of blurred objects. This rough segmentation of the blurred regions is based on an analysis of local orientation map. Then, parameters of blur are identified based on a linear constant-velocity motion blur model for every detected blurred region. After the blur parameters are estimated, the locations of blurred objects can be refined before going to a restoration process by using information from the identified blur parameters. Blur locations are refined by observing local power of the blurred image which is filtered by a high-pass filter. The high-pass filter has approximately a frequency characteristic that is complementary to the identified blur point spread function. As a final step, the image is restored by using the estimated blur parameters and locations based on an iterative deconvolution scheme applied with a regularization concept. Experimental examples of simulated and real world blurred images are demonstrated to confirm the performance of the proposed scheme.

In this paper, we present an efficient index structure for NAND flash memory, called the Dynamic Forest (D-Forest). Since write operations incur high overhead on NAND flash memory, D-Forest is designed to minimize write operations for index updates. The experimental results show that D-Forest significantly reduces write operations compared to the conventional B+-tree.

A novel error concealment method using a Kalman filter is presented in this paper. In order to successfully utilize the Kalman filter, its state transition and observation models that are suitable for the video error concealment are newly defined as follows. The state transition model represents the video decoding process by a motion-compensated prediction. Furthermore, the new observation model that represents an image blurring process is defined, and calculation of the Kalman gain becomes possible. The problem of the traditional methods is solved by using the Kalman filter in the proposed method, and accurate reconstruction of corrupted video frames is achieved. Consequently, an effective error concealment method using the Kalman filter is realized. Experimental results showed that the proposed method has better performance than that of traditional methods.

With quick topology changes due to mobile node movement or signal fading in MANET environments, conventional routing restoration processes are costly to implement and may incur high flooding of network traffic overhead and long routing path latency. Adopting the traditional shortest path tree (SPT) recomputation and restoration schemes used in Internet routing protocols will not work effectively for MANET. An object of the next generation SPT restoration system is to provide a cost-effective solution using an adaptive learning control system, wherein the SPT restoration engine is able to skip over the heavy SPT computation. We proposed a novel SPT restoration scheme, called Cognitive Shortest Path Tree Restoration (CSPTR). CSPTR is designed based on the Network Simplex Method (NSM) and Sensitivity Analysis (SA) techniques to provide a comprehensive and low-cost link failure healing process. NSM is used to derive the shortest paths for each node, while the use of SA can greatly reduce the effort of unnecessary recomputation of the SPT when network topology changes. In practice, a SA range table is used to enable the learning capability of CSPTR. The performance of computing and communication overheads are compared with other two well-known schemes, such as Dijstra's algorithm and incremental OSPF. Results show that CSPTR can greatly eliminate unnecessary SPT recomputation and reduce large amounts of the flooding overheads.

Optical burst switching (OBS) is a promising approach for the realization of future flexible high-speed optical networks. In particular, a multicast forwarding method for optical bursts is important if an efficient high-speed grid computing network is to be realized. In OBS networks, the number of burst replicas generated at each node is strongly restricted due to optical power impairment of multicast bursts. Moreover, unrestricted replication of multicast bursts at each OBS node may not be advantageous because an increase in the number of multicast bursts within the network causes more frequent deflection forwarding of both multicast and unicast bursts. This paper proposes an efficient hop-by-hop multicast forwarding method for optical bursts, where idle output ports are selected based on scores simply calculated using a routing table that each OBS node holds. This method can mitigate increases in loss rate and transfer delay of multicast bursts, even if the number of burst replicas generated at each OBS node is strongly restricted. Moreover, this method can efficiently mitigate an increase in the number of multicast bursts within the network by avoiding unnecessary replication of multicast bursts at each OBS node. Simulation results show that the proposed method can actually mitigate degradation of the loss rate and transfer delay for multicast bursts under the restricted number of burst replicas at each OBS node. Moreover, when the arrival rate of multicast bursts is large relative to that of unicast bursts, the proposed method is able to improve the loss rates of both multicast and unicast bursts by switching the forwarding method for the multicast bursts to the simple unicast forwarding method without burst replication.

This paper shows that there is a fruitful world behind sampling theorems. For this purpose, the sampling problem is reformulated from a functional analytic standpoint, and is consequently revealed that the sampling problem is a kind of inverse problem. The sampling problem covers, for example, signal and image restoration including super resolution, image reconstruction from projections such as CT scanners in hospitals, and supervised learning such as learning in artificial neural networks. An optimal reconstruction operator is also given, providing the best approximation to an individual original signal without our knowing the original signal.

Image restoration based on Bayesian estimation in most previous studies has assumed that the noise accumulated in an image was independent for each pixel. However, when we take optical effects into account, it is reasonable to expect spatial correlation in the superimposed noise. In this paper, we discuss the restoration of images distorted by noise which is spatially correlated with translational symmetry in the realm of probabilistic processing. First, we assume that the original image can be produced by a Gaussian model based on only a nearest-neighbor effect and that the noise superimposed at each pixel is produced by a Gaussian model having spatial correlation characterized by translational symmetry. With this model, we can use Fourier transformation to calculate system characteristics such as the restoration error and also minimize the restoration error when the hyperparameters of the probabilistic model used in the restoration process coincides with those used in the formation process. We also discuss the characteristics of image restoration distorted by spatially correlated noise using a natural image. In addition, we estimate the hyperparameters using the maximum marginal likelihood and restore an image distorted by spatially correlated noise to evaluate this method of image restoration.

In low bit-rate video transmission, the payload of a single packet can often contain a whole coded frame due to the high compression ratio in both spatial and temporal domains. Thus, the loss of a single packet can lead to the loss of a whole video frame. In this paper, we propose a novel error concealment algorithm that can effectively reconstruct the lost frame and protect the quality of video streams from the degradation caused by propagation errors. The proposed algorithm employs a bilateral motion estimation scheme where the weighted sum of the received motion vectors (MVs) in the neighboring frames is utilized to construct the MV field for the concealed frame. Unlike the conventional algorithms, the proposed scheme does not produce any overlapped pixel and hole region in the reconstructed frame. The proposed algorithm can be applied not only to the case of single frame loss but also adaptively extended to the case of multiframe loss. Experimental results show that the proposed algorithm outperforms other conventional techniques in terms of both peak signal-to-noise ratio (PSNR) performance and subjective visual quality.

In this paper, a Kalman filter-based method for restoration of video images acquired by an in-vehicle camera in foggy conditions is proposed. In order to realize Kalman filter-based restoration, the proposed method clips local blocks from the target frame by using a sliding window and regards the intensities in each block as elements of the state variable of the Kalman filter. Furthermore, the proposed method designs the following two models for restoration of foggy images. The first one is an observation model, which represents a fog deterioration model. The proposed method automatically determines all parameters of the fog deterioration model from only the foggy images to design the observation model. The second one is a non-linear state transition model, which represents the target frame in the original video image from its previous frame based on motion vectors. By utilizing the observation and state transition models, the correlation between successive frames can be effectively utilized for restoration, and accurate restoration of images obtained in foggy conditions can be achieved. Experimental results show that the proposed method has better performance than that of the traditional method based on the fog deterioration model.

Considering the inaccuracy of image registration, we propose a new regularization restoration algorithm to solve the ill-posed super-resolution (SR) problem. Registration error is used to obtain cross-channel error information caused by inaccurate image registration. The registration error is considered as the noise mean added into the within-channel observation noise which is known as Additive White Gaussian Noise (AWGN). Based on this consideration, two constraints are regulated pixel by pixel within the framework of Miller's regularization. Regularization parameters connect the two constraints to construct a cost function. The regularization parameters are estimated adaptively in each pixel in terms of the registration error and in each observation channel in terms of the AWGN. In the iterative implementation of the proposed algorithm, sub-sampling operation and sampling aliasing in the detector model are dealt with respectively to make the restored HR image approach the original one further. The transpose of the sub-sampling operation is implemented by nearest interpolation. Simulations show that the proposed regularization algorithm can restore HR images with much sharper edges and greater SNR improvement.

Bolstering survivable backbone networks against multiple failures is becoming a common concern among telecom companies that need to continue services even when disasters occur. This paper presents a multiple-failure recovery scheme that considers the operation and management of optical paths. The presented scheme employs scheme escalation from pre-planned restoration to full rerouting. First, the survivability of this scheme against multiple failures is evaluated considering operational constraints such as route selection, resource allocation, and the recovery order of failed paths. The evaluation results show that scheme escalation provides a high level of survivability even under operational constraints, and this paper quantitatively clarifies the impact of these various operational constraints. In addition, the fundamental functions of the scheme escalation are implemented in the Generalized Multi-Protocol Label Switching control plane and verified in an optical-cross-connect-based network.

This paper proposes a novel algorithm for image feature extraction-the dual two-dimensional fuzzy class preserving projections ((2D)2FCPP). The main advantages of (2D)2FCPP over two-dimensional locality preserving projections (2DLPP) are: (1) utilizing the fuzzy assignation mechanisms to construct the weight matrix, which can improve the classification results; (2) incorporating 2DLPP and alternative 2DLPP to get a more efficient dimensionality reduction method-(2D)2LPP.

In UMTS (universal mobile telecommunications system) networks upgraded with HSPA (high speed packet access) technology, the high access bandwidth and advanced mobile devices make it applicable to share large files among mobile users by peer-to-peer applications. To receive files quickly is essential for mobile users in file sharing applications, mainly because they are subject to unstable signal strength and battery failures. While many researches present peer-to-peer file sharing architectures in mobile environments, few works focus on decreasing the time spent in disseminating files among users. In this paper, we present an efficient peer-to-peer file sharing design for HSPA networks called AFAM -- Adaptive efficient File shAring for uMts networks. AFAM can decrease the dissemination time by efficiently utilizing the upload-bandwidth of mobile nodes. It uses an adaptive rearrangement of a node's concurrent uploads, which causes the count of the node's concurrent uploads to lower while ensuring that the node's upload-bandwidth can be efficiently utilized. AFAM also uses URF -- Upload Rarest First policy for the block selection and receiver selection, which achieves real rarest-first for the spread of blocks and effectively avoids the "last-block" problem in file sharing applications. Our simulations show that, AFAM achieves much less dissemination time than other protocols including BulletPrime and a direct implementation of BitTorrent for mobile environments.