Search

A new implementation of the weight vector in a classifier based on an optical system using a volume holographic filter with a vector decomposition algorithm is proposed for high-speed data classification. Using a volume holographic filter, the correlation between the input and the decomposed weight vector is calculated for the classification score, which is the output from the decision function in the optical classifier. The authors experimentally demonstrate the core concept and confirm the effectiveness of the proposed method by executing a two-class classification problem using a volume hologram medium.

Background: The corneal back surface is known to add some against the rule astigmatism, with implications in cataract surgery with toric lens implantation. This study aimed to set up and validate a deep learning algorithm to predict corneal back surface power from the corneal front surface power and biometric measures.\ud \ud Methods: This study was based on a large dataset of IOLMaster 700 measurements from two clinical centres. N = 19,553 measurements of 19,553 eyes with valid corneal front (CFSPM) and back surface power (CBSPM) data and other biometric measures. After a vector decomposition of CFSPM and CBSPM into equivalent power and projections of astigmatism to the 0°/90° and 45°/135° axes, a multi‐output feedforward neural network was derived to predict vector components of CBSPM from CFSPM and other measurements. The predictions were compared with a multivariate linear regression model based on CFSPM components only. \ud \ud Results: After pre‐conditioning, a network with two hidden layers each having 12 neurons was derived. The dataset was split into training (70%), validation (15%) and test (15%) subsets. The prediction error (predicted corneal back surface power CBSPP – CBSPM) of the network after training and crossvalidation showed no systematic offset, narrower distributions for CBSPP – CBSPM and no trend error of CBSPP – CBSPM vs. CBSPM for any of the vector components. The multivariate linear model also showed no systematic offset, but broader distributions of the prediction error components and a systematic trend of all vector components vs. CFSPM components. \ud \ud Conclusion: The neural network approach based on CFSPM vector components and other biometric measures outperforms the multivariate linear model in predicting corneal back surface power vector components. Modern biometers can supply all parameters required for this algorithm, enabling reliable predictions for corneal back surface data where direct corneal back surface data are unavailable.

We study the problem of talking head animation from a single image. Most of the existing methods focus on generating talking heads for human. However, little attention has been paid to the creation of talking head anime. In this paper, our goal is to synthesize vivid talking heads from a single anime image. To this end, we propose cascade pose transform network, termed CPTNet, that consists of a face pose transform network and a head pose transform network. Specifically, we introduce a mask generator to animate facial expression (e.g., close eyes and open mouth) and a grid generator for head movement animation, followed by a fusion module to generate talking heads. In order to handle large motion and obtain more accurate results, we design a pose vector decomposition and cascaded refinement strategy. In addition, we create an anime talking head dataset, that includes various anime characters and poses, to train our model. Extensive experiments on our dataset demonstrate that our model outperforms other methods, generating more accurate and vivid talking heads from a single anime image.

Rapid advancement in the communication sector through mobile networks has made handheld devices such as smartphones and tabs as a medium for most of their private transactions and communications. This necessitates the speed and efficiency of communication at a lower cost, without compromising the security. The proposed scheme is an authenticated key agreement protocol for the communication between the user and the service provider. This scheme supports batch verification and reduces the number of verification steps required and is based on the hardness of the vector decomposition problem (VDP).

We propose a new multiple image encryption scheme based on a pixel exchange operation and a basic vector decomposition in Fourier domain. In this algorithm, original images are imported via a pixel exchange operator, from which scrambled images and pixel position matrices are obtained. Scrambled images encrypted into phase information are imported using the proposed algorithm and phase keys are obtained from the difference between scrambled images and synthesized vectors in a charge-coupled device (CCD) plane. The final synthesized vector is used as an input in a random phase encoding (DRPE) scheme. In the proposed encryption scheme, pixel position matrices and phase keys serve as additional private keys to enhance the security of the cryptosystem which is based on a 4-f system. Numerical simulations are presented to demonstrate the feasibility and robustness of the proposed encryption scheme.

A high level of expertise, rigorous algorithms and methods are needed to adequately mine and harness Mass Spectrometer generated data due to its unique nature and structure. Hitherto, peptide ions are matched with theoretical results and/or public databases in order to identify expressed proteins in analyzed protein source samples, but this is done on a spectrum by spectrum basis. In this study, we present a mechanism that extends the principle of K-nearest neighbor algorithm for mining pools of mass spectrometer saliva data towards discovering and characterizing patterns for diagnosing Alzheimer's disease. The methodology discusses feature selection by correlation matrix, matrix to vector decomposition, an extension of Euclidean distance formula, and successfully classifies donor samples into the three stages of Alzheimer's disease with over 85% accuracy without collaborating clinical records.

Active thermography provides infrared images that contain sub-surface defect information, while visible images only reveal surface information. Mapping in- frared information to visible images o ers more comprehensive visualization for decision-making in rail inspection. However, the common information for regis- tration is limited due to di erent modalities in both local and global level. For example, rail track which has low temperature contrast reveals rich details in visible images, but turns blurry in the infrared counterparts. This paper pro- poses a registration algorithm called Edge-Guided Speeded-Up-Robust-Features (EG-SURF) to address this issue. Rather than sequentially integrating local and global information in matching stage which su ered from buckets e ect, this al- gorithm adaptively integrates local and global information into a descriptor to gather more common information before matching. This adaptability consists of two facets, an adaptable weighting factor between local and global information, and an adaptable main direction accuracy. The local information is extracted using SURF while the global information is represented by shape context from edges. Meanwhile, in shape context generation process, edges are weighted ac- cording to local scale and decomposed into bins using a vector decomposition manner to provide more accurate descriptor. The proposed algorithm is qual- itatively and quantitatively validated using eddy current pulsed thermography scene in the experiments. In comparison with other algorithms, better perfor- mance has been achieved. This work is supported by Engineering and Physical Sciences Research Coun- cil (EPSRC) [grant numbers EP/K503885/1]. Chaoqing Tang would also like to thank China Scholarship Council for funding his Ph.D study.

Positioning control of an underwater robot is a challenging problem due to the high disturbances of ocean flow. To overcome the high disturbance, a new underwater robot with tilting thrusters was proposed previously, which can compensate for disturbance by focusing the thrusting force in the direction of the disturbance. However, the tilting motion of the thrusters makes the system nonlinear, and the limited tilting speed sometimes makes the robot unstable. Therefore, an optimized controller is necessary. A new positioning controller is proposed for this robot using a vector decomposition method. Based on the dynamic model, the nonlinear force input term of the tilting thrusters is decomposed in the horizontal and vertical directions. Based on the decomposition, the solution is determined by a pseudo-inverse and null-space solution. Using the characteristics of the decomposed input matrix, the final solution can be found by solving a simple second-order algebraic equation to overcome the limitations of the tilting speed. The positioning was simulated to validate the proposed controller by comparing the results with a switching-based controller. Tracking results are also presented. In future work, a high-level control strategy will be developed to take advantage of the tilting thrusters by focusing the forcing direction toward the disturbance with a limited stability margin.

In this paper, the security of optical cryptosystems based on the vector decomposition technique in the Fourier domain is analyzed. Compared to the conventional cryptosystem based on the equal modulus decomposition (EMD) technique, an additional EMD structure is introduced in the cascaded EMD-based cryptosystem; hence, the mask including the phase information of the Fourier spectrum is further encoded in the second EMD structure to enhance the security level. However, it is shown that the number of the private keys has not been increased in the cascaded EMD-based cryptosystem, which makes it possible to crack the cascaded EMD-based cryptosystem. Therefore, a chosen-plaintext attack (CPA) and a special attack with an arbitrarily given private key are proposed to retrieve information from encoded images obtained by the cascaded EMD-based cryptosystem. In addition, the security of the cryptosystem based on the random modulus decomposition (RMD) technique is also analyzed. Compared to the EMD-based cryptosystem in which the Fourier spectrum is decomposed into two vectors with equal moduli, the security level of the cryptosystem has been improved by using the RMD technique to decompose the spectrum into vectors with unequal moduli to decrease the number of the amplitude constraints. However, it is found that the arbitrarily given ciphertext provides the attackers enough information to retrieve the precise information of the plaintext without any knowledge of the private keys. A special attack is proposed to crack the RMD-based cryptosystem. This is the first time to report that these two cryptosystems based on the vector decomposition technique are attacked successfully. Numerical simulation is conducted to validate the feasibility and effectiveness of the proposed attacks.

In the image quality assessment (IQA) research field, the structural similarity index measurement (SSIM) method and human visual system (HVS) model have received much attention. However, this paper shows that the definition of the luminance comparison function in SSIM conflicts with the Weber-Fechner law in HVS. To reconcile this contradiction, we propose a space similarity decomposition model based on the Weber-Fechner law and SSIM's structure parameter to decompose the image into mean value, similarity, and vertical components. Experimental results show that the vertical component is the main factor of subjective perception and should be used by itself to assess image quality. This method not only can be used to independently assess image quality, it can also be combined with other IQA methods as a preprocessing step. To solve the conflict between structural similarity method (SSIM) and Weber-Fechner law that derived from human visual system (HVS).Experiment shows that three components have different visual perception and vertical component is the main factor in subjective perception, so it should be used to assess image quality only.This space similarity decomposition model not only can be used to assess image quality independently, it could also be used as a preprocessing step in combination with other IQA schemes.

This paper presents a novel application of compositional data analysis methods in the context of color image processing. A vector decomposition method is proposed to reveal compositional components of any vector with positive components followed by compositional data analysis to demonstrate the relation between color space concepts such as hue and saturation to their compositional counterparts. The proposed methods are applied to a magnetic resonance imaging dataset acquired from a living human brain and a digital color photograph to perform image fusion. Potential future applications in magnetic resonance imaging are mentioned and the benefits/disadvantages of the proposed methods are discussed in terms of color image processing., Comment: 13 pages, 3 figures, short paper, submitted to Austrian Journal of Statistics compositional data analysis special issue, first revision, fix rendering error in fig2

The single value decomposition technique divides the data of different parties during the process of privacy preservation. The process of single value decomposition implied in the form of clustering and classification. The combined process of clustering and classification called prototype mode for sharing privacy preservation. The utility of vector decomposition in this model is selection of data in different parties for the process of maintain the raw information. In this paper proposed a prototype model for privacy preservation and improved the efficiency of data utility and accuracy of data recovery during the process of privacy. The proposed model implements in MATLAB software and used some standard dataset for evaluation of performance. The proposed model is very efficient in compression of KPPDM model. General Terms Privacy preservation, Data mining, Vector Decomposition

In the recent past, remotely sensed data with high spectral resolution has been made available and has been explored for various agricultural and geological applications. While these spectral signatures of the objects of interest provide important clues, the relatively poor spatial resolution of these hyperspectral images limits their utility and performance. In this context, hyperspectral image enhancement using multispectral data has been actively pursued to improve spatial resolution of such imageries and thus enhancing its use for classification and composition analysis in various applications. But, this also poses a challenge in terms of managing the trade-off between improved spatial detail and the distortion of spectral signatures in these fused outcomes. This paper proposes a strategy of using vector decomposition, as a model to transfer the spatial detail from relatively higher resolution data, in association with sensor simulation to generate a fused hyperspectral image while preserving the inter band spectral variability. The results of this approach demonstrates that the spectral separation between classes has been better captured and thus helped improve classification accuracies over mixed pixels of the original low resolution hyperspectral data. In addition, the quantitative analysis using a rank-correlation metric shows the appropriateness of the proposed method over the other known approaches with regard to preserving the spectral signatures.

Image classification involves extraction of repeatable and robust feature points from images for classification. These extracted feature points are described in terms of feature vectors. Scale Invariant Feature Transform (SIFT) is one of the popular feature descriptor for obtaining the feature vectors from images for image classification, image retrieval and object recognition. However, the dimension of feature vector obtained using SIFT is high. Hence, the computation time to build decision model using SIFT for image classification, image retrieval and object recognition is high and also requires large memory. In this work, we have proposed SIFT-64 and SIFT-32 to reduce the dimension of the conventional SIFT for image classification. To check the efficacy of the proposed SIFT-64 and SIFT-32, experiments were performed on two well-known publicly available datasets namely, CalTech6 and Graz. The performance is evaluated in terms of classification accuracy, training and testing time. Experimental results demonstrate superior performance of the SIFT-64 and SIFT-32 in comparison to conventional SIFT in terms of both training and testing time without compromising much on classification accuracy. The proposed feature descriptor outperforms the PCA-SIFT descriptor in terms of classification accuracy, training and testing time.

This paper presents a method aimed at encrypting color images using vector operations and secondary image phase masks (SPMs). In this encryption, the optical coherent superposition principle is used to separate and overlay the original image into two phase plates on the three channels R, G and B, respectively. Then the double random phase encoding in the Fresnel domain is performed using the SPMs to encrypt one of the phase plates. Finally, the Kronecker product of two random matrices is directly perform randomization on the encoded image, achieving multilevel encryption of color images. Computer simulations are presented to demonstrate its performance, and the feasibility and security of the proposed system are analyzed as well.

In this paper, the precision test method of single-axis PV track device is proposed by analyzing generator mechanism of PV module. In the process of the rotation of single-axis track device, the change of corner in space is considered. Suitable coordinate system is built in order to evaluate track device precision which normal vector of PV module and solar direction vector can be decomposed in this coordinate system. Meanwhile, the comparison on the tracking precision of PV track devices tested by different devices is made and the test uncertainty is researched as well. Experimental results verify the effectiveness of the method which is proposed in this paper.

International audience; In this work, a numerical finite element model of a 1.82 kg bird impacting a sandwich shield at 175 m/s is developed. Different shield designs are simulated and it appears that very different sandwich behaviours can occur, depending on the design. A new tool to analyse the deformation of the sandwich during impact is presented and is used to study the behaviour of a shield. As this tool makes it possible to easily compare the behaviour of different shields, it is used in a screening study to identify the more influential sandwich design parameters. If all design parameters are considered to be independent, the core out-of-plane plastic plateaus appear to be the most important. The core in-plane properties, elastic modulus and density and the back skin thickness have much less influence on the sandwich deformation under impact.

The present paper proposes an unsupervised feature learning method for POLSAR image classification. The proposed method includes two steps. In these two steps, features are created and learned from scratch. The first is to learn dictionaries and encode features using scatter matrices. The dictionary is learned using a set of vectors that are known as hierarchical matching pursuit (HMP). The dictionary is learned with K-singular vector decomposition (K-SVD). Afterward, the sparse codes can be computed with orthogonal matching pursuit (OMP). The second step extracts features from the previous step. The results demonstrate that the features extracted and learned from this method led to more efficient POLSAR classification results than other existing similar methods.

Truncated feature representation for automatic target detection using transformed data -based decomposition Vahid R. Riasati, Ph.D . Raytheon, Spa ce and Airborne Systems El Segundo, CA 90245 In this work, the data covariance matrix is diagonalized to provide an orthogonal bases set using the eigen vectors of the data. The eigen -vector decomposition of the data is transformed and filtered in the transform domain to truncate the data for robust feat ures related to a specified set of targets. These truncated eigen features are then combined and reconstructed to utilize in a composite filter and consequently utilized for the automatic target detection of the same class of targets. The results associated with the testing of the current technique are evaluated using the peak -correlation and peak -correlation energy metrics and are presented in this work. The inverse transformed eigen -bases of the current technique may be thought of as an injected sparsity to minimize data in representing the skeletal data structure information associated with the set of targets under consideration. Key Words: Eigen -vector/value Sampling, Sparse sampling, PSDF, Projection -Slice Theorem, Object Classification, Compressive Sensing, Pattern Recognition, Total Variation

In the context of parameter estimation, subspace-based methods like ESPRIT have become common. They require a subspace separation e.g. based on eigenvalue/-vector decomposition. In time-varying environments, this can be done by subspace trackers. One class of these is based on the PAST algorithm. Our non-linear parameter estimation algorithm DaPT builds on-top of the ESPRIT algorithm. Evaluation of the different variants of the PAST algorithm shows which variant of the PAST algorithm is worthwhile in the context of frequency estimation.

Reverse time migration (RTM) was implemented with a modified crosscorrelation imaging condition for data from 2D elastic vertically transversely isotropy (VTI) media. The computation cost was reduced because scalar qP- and qS-wavefield separations are performed in VTI media, for the source and receiver wavefields only at the RTM imaging time, to calculate the migrated qP and qS images. Angle-domain common-image gathers (CIGs) were extracted from qPqP and qPqS common-source RTM images. The local incident angle was produced as the difference between the qP-wave phase angle, obtained directly from the source wavefield polarization, and the normal to the reflector, calculated as the instantaneous wavenumber direction via a directional Hilbert transform of the stacked image. Angle-domain CIGs were extracted by reordering the prestack-migrated images by local incident phase angle, source by source. Vector decomposition of the source qP-wavefield was required to calculate the qP-wave phase polarization direction for each image point at its imaging time. RTM and angle-domain CIG extraction were successfully implemented and illustrated with a synthetic 2D elastic VTI example.

We propose an optical single-channel color image cryptosystem based on vector decomposition and three-dimensional (3D) chaotic maps. In the proposed cryptosystem, the color image is first decomposed into R, G, and B channels, and then the three channels are individually scrambled by a generalized Arnold transform. Subsequently, the scrambled channels are phase encoded and then synthesized into one vector. The synthesized vector is used as the input image of the double random phase encoding (DRPE) in the gyrator domain. The introduction of the vector decomposition enables the cryptosystem to realize single-channel color image encryption, and the first phase mask of DRPE acts as the main secret key when the input image is a non-negative amplitude-only map. In addition, the phase masks of DRPE are generated by a 3D logistic map, which can facilitate the update, management, and transmission of the phase keys. Numerical simulation results demonstrate that the proposed cryptosystem is robust against various attacks and outperforms the other relevant cryptosystems. Furthermore, the proposed cryptosystem can be directly extended to encrypt multiple color images.

A simple circuital approach to evaluate the shielding effectiveness (SE) of rectangular enclosures with apertures is proposed, considering oblique incidence and polarization. The scope of the proposed model is extended beyond 1 GHz, including higher order modes of the cavity. Furthermore, apertures are not required to be on the front face of the enclosure. In the proposed model, the SE of the enclosure with apertures on multiple sides can be simply calculated by vector decomposition. The proposed model has been successfully verified by using a conventional full-wave simulation tool. We also measured the SE of a manufactured aluminum enclosure with apertures on multiple sides.

In this paper we attempt to distinguish the direct effect of financial development on poverty reduction from its indirect effect through economic growth. Using an efficient estimator called fixed effect vector decomposition (FEVD) we employ a set of panel data from 54 developing countries for the period 1993–2004. Our results indicate that on average financial development is conducive for poverty reduction but the instability accompanying financial development is detrimental to the poor. The major policy recommendations suggested by the paper indicate that financial sector reforms should be directed at easing credit restrictions while taking into consideration the effects of financial instability on the poor.

In the interference-based encryption system, inherent silhouette problem and precise alignment problem are resolved by pixel vector decomposition algorithm and tiling technique respectively. Experimental results are presented.

Linear Discriminant Analysis (LDA) is a widely used technique for pattern classification. It seeks the linear projection of the data to a low dimensional subspace where the data features can be modelled with maximal discriminative power. The main computation in LDA is the dot product between LDA base vector and the data point which involves costly element-wise floating point multiplications. In this paper, we present a fast linear discriminant analysis method called binary LDA (B-LDA), which possesses the desirable property that the subspace projection operation can be computed very efficiently. We investigate the LDA guided non-orthogonal binary subspace method to find the binary LDA bases, each of which is a linear combination of, a small number of Haar-like box functions. We also show that B-LDA base vectors are nearly orthogonal to each other. As a result, in the non-orthogonal vector decomposition process, the computationally intensive pseudo-inverse projection operator can be approximated by the direct dot product without causing significant distance distortion. This direct dot product projection can be computed as a linear combination of the dot products with a small number of Haar-like box functions which can be efficiently evaluated using the integral image. The proposed approach is applied to face recognition on ORL and FERET dataset. Experiments show that the discriminative power of binary LDA is preserved and the projection computation is significantly reduced. (c) 2007 Elsevier B.V. All rights reserved.

In this current era, cloud environment has become essential for large-scale data storage which makes it challenging to control and prevent unauthorized access to the data. At the same time trusting the cloud provider’s integrity is also significant. Instead of assuming that the cloud providers are trusted, data owners can store the encrypted data. In 2013, Zhou et al. put forward a role-based encryption scheme (RBE) that integrates encryption with role-based access control (RBAC). In 2008, Okamoto and Takashima introduced a higher dimensional vector decomposition problem (VDP) and a trapdoor bijective function. In this paper, we propose a RBE scheme using VDP.

A reference frame is required to specify how motion is perceived. For example, the motion of part of an object is usually perceived relative to the motion of the object itself. Johansson (Psychological Research, 38, 379-393, 1976) proposed that the perceptual system carries out a vector decomposition, which rewsults in common and relative motion percepts. Because vector decomposition is an ill-posed problem, several studies have introduced constraints by means of which the number of solutions can be substantially reduced. Here, we have adopted an alternative approach and studied how, rather than why, a subset of solutions is selected by the visual system. We propose that each retinotopic motion vector creates a reference-frame field in the retinotopic space, and that the fields created by different motion vectors interact in order to determine a motion vector that will serve as the reference frame at a given point and time in space. To test this theory, we performed a set of psychophysical experiments. The field-like influence of motion-based reference frames was manifested by increased nonspatiotopic percepts of the backward motion of a target square with decreasing distance from a drifting grating. We then sought to determine whether these field-like effects of motion-based reference frames can also be extended to stationary landmarks. The results suggest that reference-field interactions occur only between motion-generated fields. Finally, we investigated whether and how different reference fields interact with each other, and found that different reference-field interactions are nonlinear and depend on how the motion vectors are grouped. These findings are discussed from the perspective of the reference-frame metric field (RFMF) theory, according to which perceptual grouping operations play a central and essential role in determining the prevailing reference frames.

Ultrasound waves interact strongly with the orientation and sequence of the plies in a layup when propagating in the thickness direction of composite laminates. Also the layup orientation greatly influences its properties in a composite laminate. If the layup orientation of a ply is misaligned, it could result in the part being rejected and discarded. Now, most researchers cut a small coupon from the waste edge and use a microscope to optically verify the ply sequences on important parts. This may add a substantial cost to the production since the test is both labor intensive and performed after the part is cured. A nondestructive technique would be very beneficial, which could be used to test the part after curing and requires less time than the optical test. Therefore we have developed, reduced, and implemented a novel ply-by-ply vector decomposition model for composite laminates fabricated from unidirectional plies. This model decomposes the transmission of a linearly polarized ultrasound wave into orthogonal components through each ply of a laminate. High probability is found, by comparisons between the model and tests, in characterizing cured layups of the laminates by using the proposed method.

Oblivious transfer is one of the most basic and widely used protocol primitives in cryptography. It can be described as a two-party protocol used for interaction between a sender and a receiver. A 1-out-of-2 oblivious transfer is the interaction between a sender and a receiver in which a sender has two strings m 0 and m 1. At the end of the interaction, receiver learns exactly one of the strings m 0 and m 1, while the sender learns nothing. Lindell showed how to achieve efficient and fully simulatable non-adaptive oblivious transfer under decisional Diffie–Hellman (DDH) problem, Nth residuosity and quadratic residuosity assumptions, as well as the assumption that homomorphic encryption exists. We propose a scheme based on this protocol under the assumption namely vector decomposition problem. Our scheme is non-adaptive and fully simulatable.

Purpose To test several methods of assessing surgically induced astigmatism (SIA). Setting Department of Ophthalmology, Arhus University Hospital, Arhus, Denmark. Methods Assessment methods can be divided into 3 groups. Group 1 includes methods that consider only astigmatic magnitude and disregard astigmatic direction, such as the algebraic and simple subtraction methods. Group 2 can be termed “astigmatic magnitude not considering axis” and includes several almost identical techniques known as the methods of Naylor, Jaffe, Kaye, Holladay, and Olsen. Group 3 includes Naeser polar value analysis and the methods later described by Alpins and Holladay. The methods were tested in situations in which the expected result was known. When this result was not produced, the specific assessment method was considered falsified in a classical Popperian manner. Results The simple falsification experiments revealed that the methods in Groups 1 and 2 are erroneous. Mathematical analysis disclosed that the methods in Group 3 are similar, although derived from different concepts. Conclusion The algebraic, simple subtraction, astigmatic magnitude not considering axis, and vector decomposition methods for assessment of SIA are erroneous and should not be used. The Naeser, Alpins, and Holladay methods are identical and in agreement with current research in optometry.

There has been much discussion in the literature about rival measures of classical polarization in three dimensions. We gather and compare the various proposed measures of polarization, creating a geometric representation of the polarization state space in the process. We use majorization, previously used in quantum information, as a criterion to establish a partial ordering on the polarization state space. Using this criterion and other considerations, the most useful polarization measure in three dimensions is found to be one dependent on the Bloch vector decomposition of the polarization matrix., 8 pages

Iterative deconvolution is an effective method of achieving high azimuth resolution in real aperture scanning radar. However, the performance is seriously limited when SNR is low. The application area and processing efficiency are affected by unreasonable iterative. Therefore, correct understanding the ability of super-resolution signal recovery is the premise to achieve super-resolution at low SNR. In this paper, the distance and position change of the recovered signals in high-dimension are studied based on the vector decomposition of azimuth echo. Besides, the process of signal recovery and the limit of super-resolution performance in iteration are analyzed. The experiential formula of the super-resolution limit is then derived. Finally, simulations are given to verify the theoretical analysis.

Encryption using vector decomposition problem (VDP) on higher dimensional vector spaces is a novel method in cryptography. Yoshida has shown that the VDP on a two-dimensional vector space is at least as hard as the computational Diffie-Hellman problem on a one-dimensional subspace under certain conditions. Steven Galbraith has shown that for certain curves, the VDP is at most as hard as the discrete logarithm problem on a one-dimensional subspace. Okomoto and Takashima proposed encryption scheme and signature schemes using VDP. An authenticated key agreement scheme using vector decomposition problem is proposed in this paper

The technique of vector space decomposition control of voltage source inverter fed dual three-phase induction machines is presented in this paper. By vector space decomposition, the analytical modeling and control of the machine are accomplished in three two-dimensional orthogonal subspaces and the dynamics of the electromechanical energy conversion related and the nonelectromechanical energy conversion related machine variables are thereby totally decoupled. A space vector PWM technique is also developed based on the vector space decomposition to limit the 5th, 7th, 17th, 19th,... harmonic currents which in such a system would be otherwise difficult to control. The techniques developed in this paper can be generalized for the control of an induction machine with an arbitrary number of phases. >

In this article, a learning-based target decomposition method based on Kernel K-singular vector decomposition (Kernel KSVD) algorithm is proposed for polarimetric synthetic aperture radar (PolSAR) image classification. With new methods offering increased resolution, more details (structures and objects) could be exploited in the SAR images, thus invalidating the traditional decompositions based on specific scattering mechanisms offering low-resolution SAR image classification. Instead of adopting fixed bases corresponding to the known scattering mechanisms, we propose a learning-based decomposition method for generating adaptive bases and developing a nonlinear extension of the KSVD algorithm in a nonlinear feature space, called as Kernel KSVD. It is an iterative method that alternates between sparse coding in the kernel feature space based on the nonlinear dictionary and a process of updating each atom in the dictionary. The Kernel KSVD-based decomposition not only generates a stable and adaptive representation of the images but also establishes a curvilinear coordinate that goes along the flow of nonlinear polarimetric features. This proposed approach was verified on two sets of SAR data and found to outperform traditional decompositions based on scattering mechanisms.

A mathematical model of the three-level PWM inverter is set up. The dual close-loop control of DC motor is applied to three-level PWM inverter. As a result, the three-level inverter has favorable dynamic and steady state performance, sinusoidal input current and unity power factor. A simple three-level SVPWM algorithm with reference voltage vector decomposition presents in this paper. A neutral-point balancing control method is employed, by adjusting the times of positive small vectors and negative small vectors. Finally, computer simulations show the superiority of the method.

Summary Introduction On account of its specific biodynamics, the disc joint located at the very heart of the joint can impact every constituent of the manducatory system. The disc is deformed when subjected to stresses exerted by the muscles of mastication which it partly absorbs and partly redistributes. Materials and methods CT-scan slices and MRI images of a subject were made in order to create a finite element anatomical model of the TMJ. The forces applied to the subject's joint model were obtained by performing vector decomposition of the maximum muscle forces produced by this individual. The resultant force in this study was subjected to different frequencies approximating those observed in mastication. Results The reaction force at the glenoid fossa can reach up to 1035 N depending on the frequency of the indentation. Generally, during the different exercises, the areas of maximum stress were located at the lateral portion of the disc and on the posterior band. They reached forces up to 13.2 MPa following a 32 s exercise at a frequency of 0.5 Hz. Discussion Even if the behavior law needs to be improved, joint resiliency was demonstrated in this study. The areas of maximum stress were equivalent in the different exercises on account of the anatomy of the different parts and the axis of the forces applied. This study offers food for thought regarding joint disorders and opens the way for further research to complement the current investigation.

OBJECTIVE: To study the 1-year course of the surgically-induced astigmatism after phacoemulsification through a 6 to 6.5 mm scleral tunnel incision by using six different analytic techniques. STUDY DESIGN: Prospective study with follow-up at 1 day, 1 week, 1 month, 4 months and 1 year after surgery. SETTING: Department of Ophthalmology, Aarhus University Hospital. PATIENTS: A consecutive series of 110 patients between 22-81 years of age (mean age 60) scheduled for phacoemulsification between October 1991 and May 1992. MAIN OUTCOME MEASURES: The surgically induced astigmatism was evaluated using 6 different methods: by (1) the subtraction method, (2) vector analysis, (3) vector decomposition, (4) Cravy's vertical vector, (5) Naeser's polar values, and (6) the algebraic method, respectively. RESULTS: By subtraction, without regard to axis, the induced astigmatism was found to reach baseline level 1 month after surgery. By vector analysis, the mean induced astigmatism was 1.33 D,1.31 D, 0.75 D, 0.72 D and 0.75 D 1 day, 1 week, 1 month, 4 months and 1 year after surgery, respectively. By vector decomposition, 27%, 25%, 55%, 85% and 86% of the surgically-induced astigmatism was found to be against-the-rule at the same postoperative visits, respectively. By Cravy's method, the mean induced cylinder was +0.75, +0.89, +0.08, -0.44D and -0.50D, respectively. Similar values were found with Naeser's method and with the algebraic method. CONCLUSIONS: We conclude the surgically induced astigmatism to be stable 4 months after phacoemulsification with a 6 to 6.5 mm scleral tunnel incision. Regarding the six analytic techniques we recommend the vector analysis, which is capable of reporting both the magnitude and the axis of the induced astigmatism within the theory of Gaussian optics. To illustrate the with-the-rule and against-the-rule component in a simple manner, we suggest the vector decomposition to be used as an adjunct to the vector analysis.

The vector transform has previously been proposed for vector quantization image coding. It is intended for decorrelating intervector dependency and preserving intravector dependency. The present authors study optimally such a transform. It is found that the optimal vector transform in the above senses, with the first-order Gaussian Markov model, does not exist unless the overall correlation matrix has an intervector and intravector decomposition structure. This is especially true for vectors formed by blocks of pixels. Then, a necessary and sufficient condition is derived for general vectors on the existence of their optimal vector transforms. >

Based on the Weber-Fechner Law and decomposition method of image space vector, a space similarity analysis model of Weber-Fechner Law is proposed by analyzing space vector structure similarity. We explore the vision feeling of background brightness, contrast similarity and independence component, and establish new measurement of image quality. It is found that space model is more matching with subjective vision feeling and lowers complex and uncertain of image quality assessment from experimentation result. The conclusion is valuable for setting up new image quality assessment.

In this paper we present several algorithms for reconstructing 2D convex sets given support line measurements for which the angles are known precisely but the lateral displacements are noisy. We extend the algorithms given in a previous paper by explicitly incorporating prior information about the shape of the objects to be reconstructed. We develop the Scale-Invariant algorithms, which incorporate prior shape information by defining prior probabilities on support vectors, where a support vector is a vector formed from the lateral displacements of a particular set of support lines of an object. We also develop the Ellipse-Based algorithms, which either assume or jointly estimate the parameters of an ellipse, given prior distributions that favor ellipses. In order to relate the support vector prior probability to the expected shape of an object we develop a vector decomposition called the Size/Shape/Shift decomposition, which helps to provide insight into the detailed geometric relationship between support vectors and 2D convex objects. We then use the maximum a posteriori criterion to determine the specific form of the support vector estimator. The computations involve a quadratic programming optimization stage, which is used to determine one component of the decomposition, and either a line search or a conjugate gradient stage, which is used to determine the remaining components. The performance of the algorithms is demonstrated using simulated support line measurements of an ellipse.

The efficient and compact representation of images is a fundamental problem in computer vision. In this paper, we propose methods that use Haar- like binary box functions to represent a single image or a set of images. A desirable property of these box functions is that their inner product operation with an image can be computed very efficiently. We propose two closely related novel subspace methods to model images: The nonorthogonal binary subspace ( NBS) method and the binary principal component analysis ( B-PCA) algorithm. NBS is spanned directly by binary box functions and can be used for image representation, fast template matching, and many other vision applications. B- PCA is a structure subspace that inherits the merits of both NBS ( fast computation) and PCA ( modeling data structure information). B- PCA base vectors are obtained by a novel PCA- guided NBS method. We also show that B- PCA base vectors are nearly orthogonal to each other. As a result, in the nonorthogonal vector decomposition process, the computationally intensive pseudoinverse projection operator can be approximated by the direct dot product without causing significant distance distortion. Experiments on real image data sets show a promising performance in image matching, reconstruction, and recognition tasks with significant speed improvement.

Efficient and compact representation of images is a fundamental problem in computer vision. Principal Component Analysis (PCA) has been widely used for image representation and has been successfully applied to many computer vision algorithms. In this paper, we propose a method that uses Haar-like binary box functions to span a subspace which approximates the PCA subspace. The proposed method can perform vector dot product very efficiently using integral image. We also show that B-PCA base vectors are nearly orthogonal to each other. As a result, in the non-orthogonal vector decomposition process, the expensive pseudo-inverse projection operator can be approximated by the direct dot product without causing significant distance distortion. Experiments on real image datasets show that B-PCA has comparable performance to PCA in image reconstruction and recognition tasks with significant speed improvement.

A new vector quantization method is proposed which can reduce search complexity and code book memory size by reducing the number of code vectors without increasing quantization distortion. This method uses hyper-columnar clusters, and an input vector is quantized to a cluster, center axis of which is nearest to the input vector. Thus, one vector and one scalar must be coded and transmitted. The proposed method was applied to four geometrically different distributions and LPC cepstrums of speech signals. As a result, the number of code vectors was decreased compared with that in an ordinary vector quantizer, in all of the above distributions. Then the reduction of memory size and search complexity were evaluated. >

The knowledge of the particle size distribution of the pulverized coal used to fuel a fossil fuel power station is an important and useful parameter in the control and efficiency of power generation. However, to put this parameter to its most effective use, it is necessary to measure it on-line. An investigation conducted into techniques used for the on-line measurement of particle size distribution revealed a cost effective technique to achieve this goal. In short, it entails the measurement of the resonant vibrations or modes evoked in a structure when pneumatically conveyed particles collide with it. By analyzing the frequency spectra of these evoked vibrations or modes, variations in the size distribution of the particles colliding with the structure can be observed. The method is based on the principle that particles colliding with a structure cause that structure to start vibrating at its natural or resonant frequencies. Different sized particles cause different resonant frequencies of the structure to be exited or evoked. Thus, a variation in particle size distribution, will cause a variation in the amplitudes of the different resonant vibration peaks of the structure. By analyzing these peaks, variations in the particle size distribution can be monitored. Experimental testing and evaluation of the technique on a pilot plant proved it to be a very cost effective solution to the measurement problem. Extraction of the absolute particle size information from the complex vibration signals was accomplished using special parameter back-calculation techniques. A numerical/model-oriented approach using vector decomposition and a knowledge-based approach using neural networks allowed absolute particle size distribution to be determined.