The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Relation learning is a fundamental operation in many computer vision tasks. Recently, high-order Boltzmann machine and its variants have exhibited the great power of modelling various data relation. However, most of them are unsupervised learning models which are not very discriminative and thus cannot server as a standalone solution to relation learning tasks. In this paper, we explore supervised...
This paper presents a data-driven matching cost for stereo matching. A novel deep visual correspondence embedding model is trained via Convolutional Neural Network on a large set of stereo images with ground truth disparities. This deep embedding model leverages appearance data to learn visual similarity relationships between corresponding image patches, and explicitly maps intensity values into an...
While feedforward deep convolutional neural networks (CNNs) have been a great success in computer vision, it is important to note that the human visual cortex generally contains more feedback than feedforward connections. In this paper, we will briefly introduce the background of feedbacks in the human visual cortex, which motivates us to develop a computational feedback mechanism in deep neural networks...
Pairwise learning algorithms are a vital technique for personalized ranking with implicit feedback. They usually assume that each user is more interested in items which have been selected by the user than remaining ones. This pairwise assumption usually derives massive training pairs. To deal with such large-scale training data, the learning algorithms are usually based on stochastic gradient descent...
In this paper, we proposed control scheme combined backstepping, radius basis function neural network and adaptive control for nonlinear system with state delay which is as same as sewage treatment system, RBF neural network is used to estimate the unknown continuous function, In the backstepping design, all unknown functions at intermediate steps are passed down such that only a single neural network...
Breast cancer has been one of the most common diseases which cause many women's death throughout the world. The early breast cancer detection is detecting the tumor when it is at early stage, which could help heal the patients and would not harm the women' s health. It is clear that the best way to control this disease is detecting it at an early stage. Ultra-wide band (UWB) microwave imaging is a...
This paper presents one dynamic regulation of controller parameters strategy for power factor correction (PFC) rectifier, which is based three-phase Vienna topology. To improve the converter power-density, the operation with high frequency and fast calculation is required. In this paper, the proposed controller is based on digitized feedforward compensation approach, can improve the control system...
In this paper, we instantiate random oracle (RO) H in the Randomized Convergent Encryption proposed by Bellare et al. (Eurocrypt 2013) via a universal computation alextractor (UCE). We prove that if H is UCE secure for statistically unpredictable source, then there constructed RCE[H] is PRV$-CDA (Strong Privacy-Chosen Distribution Attacks) secure in the standard model.
Handwritten signature is widely used for human identity authentication in daily life. In order to improve the recognition accuracy and promote the success rate, it is necessary to preprocess the signature image. In this paper, the signature image preprocessing included smooth filtering, image binaryzation, signature area acquisition, and edge thinning. Furthermore, an improved adaptive median filtering...
The characteristic of the driving torque generated by the superconducting motor is analyzed. The magnetic field distribution of the stator coils and the driving torque at the different position are calculated by FEM software. According to the rotation position detected by the fiber optic sensor measurement system, the motor can drive the rotor accelerating or decelerating rotation conveniently by...
Efficiency and effectiveness are two key factors to evaluate a human segmentation algorithm for real vision applications. However, most existing algorithms only focus on one of them. That is, fast and accurate human segmentation is not yet well addressed. In this paper, we propose a super-fast and highly accurate human segmentation method with very deep convolutional neural networks. We also provide...
Recent helium re-condensing technology has been widely applied in the field of superconducting magnet. The pulse tube (PT) cryocoolers have achieved great application for its advantage of no moving parts in the cold end, high reliability, low vibration and noise. In our study, a two-stage PT cryocooler is employed to provide the cooling power for a nuclear magnetic resonance (NMR) superconducting...
The magnetic field inhomogeneity and the Lorentz force of a superconducting magnet for magnetic resonance image are optimized by changing the shape of the ferromagnetic poles. The two objectives are calculated by a commercial software Opera3D-TOSCA based on the finite element method. The genetic algorithm improved by adding niching and adaptive algorithms is developed on the computation platform MATLAB...
The high-field superconducting magnets as a core components of the MRI has been widely used in the fields of medical equipment. The superconducting magnet technology depends on the development of superconducting material properties and cryogenic technology. Compared with conventional imaging equipment, ultra-high-field magnetic resonance imaging technology has unparalleled advantages of the improvement...
An electrode-insulated electrostatically suspended gyroscope has been designed and fabricated in IEE. The four-axis non-orthogonal electrostatic suspension system has eight symmetrically aligned electrodes with insulation for the sake of safety. However, insulation of the electrodes has effect on the displacement measurement of the bearing system, the influences of which on gap capacitances and displacement...
Flow control of magnetohydrodynamic flow in a cylindrical spiral channel is studied. Magnetic field strength, voltage and magnetic-force-based rectification are used as the controllers. An axial-symmetric cylindrical spiral channel which helps to control the magnetic flow is designed. The flow is subjected to non-periodic inlet and outlet conditions where the inlet profile of the axial velocity is...
This paper presents an image processing technique to assist hypochromatopsia people with better viewing experience. We propose an image enhancement approach, called color reprint, based on a linear transformation with color space calculation to adjust the color presentation. By rearranging the color gamut, our method can help hypochromatopsia people distinguish the difference among the colors they...
Temporal dynamics of postures over time is crucial for sequence-based action recognition. Human actions can be represented by the corresponding motions of articulated skeleton. Most of the existing approaches for skeleton based action recognition model the spatial-temporal evolution of actions based on hand-crafted features. As a kind of hierarchically adaptive filter banks, Convolutional Neural Network...
We propose a novel multi-view clustering method by learning auto-regression problems under structural constraints and treating the regression coefficients as new feature representations for the cluster partition. In particular, we take the data intrinsic correlation structure into account. Correlated data under one view tend to be also related under another view and are likely to fall into the same...
Smile detection from facial images is a specialized task in facial expression analysis with many potential applications such as smiling payment, patient monitoring and photo selection. The current methods on this study are to represent face with low-level features, followed by a strong classifier. However, these manual features cannot well discover information implied in facial images for smile detection...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.