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.
It is a challenging task to detect people in crowded scene owing to the heavy occlusion and small object characteristic of crowed people. In this paper we propose a multi-layer regional-based convolutional network to address the tasks of people detection in the crowded scenes. Unlike the traditional methods, we propose an end-to-end framework that uses the convolutional network as feature extractor...
The design of improved fault location system and its experimental test system is presented in this paper. The fault location system is based on traveling waves and contains several innovative algorithms, considering the main influential factors such as the uncertainty of wave speed and the accuracy of transmission line length. Besides, the design of experimental system for testing the location precision...
In this research, the procedure of the transient traveling wave in non-solidly earthed network is analyzed. It shows that at the bus bar the traveling waves of healthy lines are mainly composed of the refracted traveling waves. The refracted are generated from the backward traveling wave in the fault line, then the refracted traveling waves is the main part of the traveling waves in the healthy lines;...
The failure rate of doubly-fed induction generators (DFIG) for wind turbine generations was relatively higher than that of direct-driven generators, according to statistical data from some wind power industries. The rotor insulation was a relatively weak part of the insulation system for DFIG because it was normally subjected to a PWM impulse voltage during the long-time operation. In this paper,...
Since the emergence of extensive multivariate data, feature fusion has been more and more important. Most conventional methods of feature fusion do not pay sufficient attention to inherent geometric properties of data, even in the case where the data have spatial features. Therefore, how to fuse multiple features in a more intuitive way is still an open problem. A new strategy of parallel graphical...
Spiral motion is a typical motion of underwater gliders, which is usually used to change the direction of motion. Compared with the sawtooth motion in the vertical plane, the mechanism of spiral motion is more complex and less understood. In order to characterize steady-state spiral motion of underwater gliders, the dynamic model for underwater gliders, steered by a single internal movable and rotatable...
This paper introduces the improved measures of Electromagnetic Compatibility (EMC) for Automatic Transmission Control Module (TCM) circuit design. The improved electromagnetic designs for TCM circuit can be divided into two aspects, which are Electromagnetic Interference (EMI) and Electromagnetic Susceptibility (EMS). The EMI characteristics are obtained by measuring the TCM. Through analyzing the...
Visual tracking is intrinsically a temporal problem. Discriminative Correlation Filters (DCF) have demonstrated excellent performance for high-speed generic visual object tracking. Built upon their seminal work, there has been a plethora of recent improvements relying on convolutional neural network (CNN) pretrained on ImageNet as a feature extractor for visual tracking. However, most of their works...
In recent years, the State Grid has built amounts of business systems, e.g. OA, marketing and MIS systems. And it gradually developed into big data. However, with the construction of information increasingly deepening and a sharp increase of data volume, new challenges and inconvenience have been brought out for data seekers. In this paper, an architecture model of data integration in power field...
Now coupling interferometer structure based on photonic crystal fiber (PCF) gets extensive attention in hydrogen sensing. Because of its structural stability, high sensitivity characteristics, it has achieved good results in the hydrogen sensor, when combine with the hydrogen-sensitive film such as Pd-WO3. In the initial experiment, the hydrogen sensor based on PCF coated with Pd-WO3 showed good linearity...
In this paper, a magnetic field sensor based on a multilongitudinal mode fiber laser has been proposed by employing a rotary apparatus modulated by an external magnetic field. Due to the remarkable birefringence change caused by the fiber twist, the polarimetric mode beat frequency (PMBF) of the laser is sensitive to the applied magnetic field intensity. Experimental results show that the PMBF linearly...
A novel hollow-core fiber (HCF) curvature sensor based on a tapered HCF sandwiched between two single mode fibers (Butterfly-Shape Structure) is proposed and experimentally demonstrated. The collapsed region around the first fusion interface excites the high-order modes, and the butterfly shape couples the high-order modes back into the core and interferes with the fundamental mode in the second fusion...
A novel refractive index (RI) sensor based on an asymmetrical Mach-Zehnder interferometer (MZI) with two different step-like tapers is proposed. The step-like taper is fabricated by fusion splicing two half tapers with an appropriate offset. By further applying offset and discharging to the last fabricated step-like taper of MZI, influence of taper parameters on interference spectrum is investigated...
Relative hovering of spacecraft is a new type of close-proximity operation. This work deals with the body-fixed hovering over a tumbling object. Initially, the dynamics of the virtual hovering point over the tumbling object in local-vertical-local-horizon(LVLH) frame is derived with considering the rotation of the object. Then, the body-fixed hovering control is converted into an optimal tracking...
Recently, the low-rank plus diagonal (LRPD) adaptation was proposed for speaker adaptation of deep neural network (DNN) models. The LRPD restructures the adaptation matrix as a superposition of a diagonal matrix and a product of two low-rank matrices. In this paper, we extend the LRPD adaptation into the subspace-based approach to further reduce the speaker-dependent (SD) footprint. We apply the extended...
The reliability performance of 28nm metal e-Fuse programmed with different current is investigated. High temperature stress (HTS) or temperature cycling (TC) may cause the shift of metal-e-fuse element resistance and shape. In this paper, we find that 28nm metal e-Fuse programming with low current reliability performance is more stable than metal e-Fuse programming with high current; Resistance shift...
Program disturbance is a major intrinsic reliability concern on NAND flash. In this paper, we present that NAND flash E/W (Erase/Write) cycle failures induced by program disturbance for erase cells VT (Threshold Voltage) positive shift. The root cause of program disturbance is the old process of gate re-oxidation issue, which results in ILD (Intra-Layer Dielectric) voids, then Ni fills in the ILD...
Hall-effect sensors are wildly utilized to estimate rotor angle of permanent magnet synchronous motor (PMSM) in low-cost situations using vector control. A variety of estimation algorithms have been proposed to improve the precision of the estimated rotor position, while few literatures so far dedicate in dealing with Hall-effect sensor faults. In this paper, the types of sensor faults are divided...
Sparse representation, which represents the test sample as a linear combination of the whole training samples, achieved great success in face recognition. It can obtain a good performance if there exist enough training samples. However, the number of face images of a subject is usually limited in real face recognition systems. In this paper, in order to obtain more representations of a face, we propose...
A new type of End-to-End system for text-dependent speaker verification is presented in this paper. Previously, using the phonetic discriminate/speaker discriminate DNN as a feature extractor for speaker verification has shown promising results. The extracted frame-level (bottleneck, posterior or d-vector) features are equally weighted and aggregated to compute an utterance-level speaker representation...
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.