National University of Defense Technology

www.nudt.edu.cn
Changsha, China

National University of Defense Technology is a comprehensive national key university based in Changsha, Hunan Province, China.It is under the dual supervision of the Ministry of National Defense and the Ministry of Education, designated for Project 211 and Project 985, the two national plans for facilitating the development of Chinese higher education. Wikipedia.


Time filter

Source Type

Liu L.,National University of Defense Technology | Fieguth P.,University of Waterloo
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2012

Inspired by theories of sparse representation and compressed sensing, this paper presents a simple, novel, yet very powerful approach for texture classification based on random projection, suitable for large texture database applications. At the feature extraction stage, a small set of random features is extracted from local image patches. The random features are embedded into a bag-of-words model to perform texture classification; thus, learning and classification are carried out in a compressed domain. The proposed unconventional random feature extraction is simple, yet by leveraging the sparse nature of texture images, our approach outperforms traditional feature extraction methods which involve careful design and complex steps. We have conducted extensive experiments on each of the CUReT, the Brodatz, and the MSRC databases, comparing the proposed approach to four state-of-the-art texture classification methods: Patch, Patch-MRF, MR8, and LBP. We show that our approach leads to significant improvements in classification accuracy and reductions in feature dimensionality. © 2012 IEEE.


Zhang J.-R.,National University of Defense Technology
Physical Review D - Particles, Fields, Gravitation and Cosmology | Year: 2013

We present a QCD sum rule analysis for the newly observed resonance X c(3250) by assuming it to be a D0*(2400)N molecular state. Technically, contributions of operators up to dimension 12 are included in the operator product expansion (OPE). We find that it is difficult to find the conventional OPE convergence in this work. By releasing the rigid OPE convergence criterion, one could find that the OPE convergence is still under control in the present work, and the numerical result for the D0*(2400)N state is 3.18±0.51 GeV, which is in agreement with the experimental data of Xc(3250). In view of that, the conventional OPE convergence is not obtained here; thus only weak conclusions can be drawn regarding the explanation of Xc(3250) in terms of a D0*(2400)N molecular state. As a byproduct, the mass for the bottom counterpart B̄0*N state is predicted to be 6.50±0.49 GeV. © 2013 American Physical Society.


Zhang J.-R.,National University of Defense Technology
Physical Review D - Particles, Fields, Gravitation and Cosmology | Year: 2013

In the framework of QCD sum rules, we present an improved study of our previous work, particularly on the D̄D* molecular state, to investigate the possibility that the newly observed Zc(3900) is a S-wave D̄D* molecular state. To ensure the quality of QCD sum rule analysis, contributions of up to dimension nine are calculated to test the convergence of operator product expansion (OPE). We find that the two-quark condensate âŸ̈q̄q⟩ is very large and makes the standard OPE convergence (i.e. the perturbative at least larger than each condensate contribution) happen at very large values of Borel parameters. By releasing the rigid OPE convergence criterion, one could find that the OPE convergence is still under control. We arrive at the numerical result 3.86±0.27 GeV for D̄D*, which agrees with the mass of Zc(3900) and could support the explanation of Zc(3900) in terms of a S-wave D̄D* molecular state. © 2013 American Physical Society.


Liu W.,National University of Defense Technology
Optics Express | Year: 2015

We study the light scattering of homogenous radially-anisotropic spherical particles. It is shown that radial anisotropy can be employed to tune effectively the electric resonances, and thus enable flexible overlapping of electric and magnetic dipoles of various numbers, which leads to unidirectional forward super-scattering at different spectral positions. We further reveal that through adjusting the radial anisotropy parameters, electric and magnetic resonances of higher orders can be also made overlapped, thus further collimating the forward scattering lobes. The ultra-directional super-scattering we have obtained with individual homogenous radially anisotropic spherical particles may shed new light to the design of compact and efficient nanoantennas, which may find various applications in solar cells, bio-sensing and many other antenna based researches. © 2015 Optical Society of America.


Gao G.,National University of Defense Technology
IEEE Geoscience and Remote Sensing Letters | Year: 2011

This letter proposes a Parzen-window-kernel-based algorithm for ship detection in synthetic aperture radar (SAR) images. First, the data-driving kernel functions of Parzen window are utilized to approximate the histogram of real SAR image, in order to complete the accurate modeling of SAR images. Then, a threshold of global constant false alarm rate is given theoretically, and the numerical solution of the threshold is also derived. The experimental results of the real data of typical targets demonstrate that the algorithm presented is effective. © 2010 IEEE.


Huang W.,National University of Defense Technology
International Journal of Hydrogen Energy | Year: 2014

The information in the three-dimensional transverse injection flow field is very important for the design of a scramjet combustor, and it should be explored by using the data mining and multi-objective design optimization methods. In the current study, the three-dimensional Reynolds-averaged Navier-Stokes (RANS) equations coupled with the two equation SST k-ω turbulence model has been utilized to simulation the transverse injection flow field with a freestream Mach number 3.75, and the influence of the turbulence model on the flow field properties has been evaluated as well. At the same time, three grid scales have been employed to perform the grid independency analysis, and the predicted results have been compared with the available experimental data in the open literature in order to carry out the code validation process. Further, the effect of the injector geometric configuration on the mixing efficiency of the transverse injection flow field has been investigated, and four different configurations have been considered in the current study, namely the square port, the diamond port, the equilateral triangular port and the circular port. The obtained results show that the case with the square injection port can obtain the largest mixing efficiency, and it can offer the rapidest near-field mixing between the injectant and the air. At last, the transverse injection flow field with the square injection port has been optimized by the surrogate-based evolutionary algorithm, and the relationships between the design variables and the objective functions have been explored by the variance analysis method. It is shown that the jet-to-crossflow pressure ratio has a high remarkable impact on the total pressure recovery efficiency, as well as the number of the injection ports on the drag force performance. The drag force increases with the increase of the number of the injection ports due to the deeper penetration of the rear jets. Copyright © 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.


Zhang J.-R.,National University of Defense Technology
Physical Review D - Particles, Fields, Gravitation and Cosmology | Year: 2014

Theoretically, some works have proposed the hadronic resonances Σc(2800) and Λc(2940)+ to be S-wave DN and D*N molecular candidates, respectively. In the framework of QCD sum rules, we investigate whether Σc(2800) and Λc(2940)+ could be explained as the S-wave DN state with JP=12- and the S-wave D*N state with JP=32-, respectively. Technically, contributions of operators up to dimension 12 are included in the operator product expansion. The final results are 3.64±0.33GeV and 3.73±0.35GeV for the S-wave DN state of JP=12- and the S-wave D*N state of JP=32-, respectively. They are somewhat bigger than the experimental data of Σc(2800) and Λc(2940)+, respectively. Since corresponding molecular currents are constructed from local operators of hadrons, the possibility of Σc(2800) and Λc(2940)+ as molecular states cannot be arbitrarily excluded merely from these disagreements between molecular masses using local currents and experimental data. But then these results imply that Σc(2800) and Λc(2940)+ could not be compact states. This may suggest a limitation of the QCD sum rule using the local current to determine whether some state is a molecular state or not. As by-products, masses for their bottom partners are predicted to be 6.97±0.34GeV for the S-wave B̄N state of JP=12- and 6.98±0.34 GeV for the S-wave B̄*N state of JP=32-. © 2014 American Physical Society.


Patent
National University of Defense Technology | Date: 2014-03-05

An automatic monitoring method of the settlement of engineering buildings is provided. A measuring station is provided at each point to be measured. The measuring stations include multi-head camera stations (11) and markers (12) which can settle with the engineering buildings. The vertical displacement of the position of each measuring station is obtained by computing relationships of the displacement transfer measurement which are built by the coordinate data of each marker (12) in accordance with camera-series imaging geometry constraint relationship, to monitor the settlement of engineering buildings. At the same time, an automatic monitoring system of the settlement of engineering buildings is provided, to achieve an automatic, longtime and continuous measurement of the settlement of engineering buildings such as roadbeds, floors, bridges, etc.


Patent
National University of Defense Technology | Date: 2015-01-15

Disclosure is a method for detecting the safety driving state of a driver, the method comprises the following steps: (a) detecting the current sight direction of a driver in real time and acquiring a scene image signal in a front view field of the driver when a vehicle runs; (b) processing the acquired current road scene image signal according to a visual attention calculation model to obtain the expected attention distribution of the driver under the current road scene; and (c) performing fusion analysis on the real-time detected current sight direction of the driver in the step (a) and the calculated expected attention distribution of the driver in step (b), and judging whether the current driver is in a normal driving state and whether the driver can timely make a proper response to the sudden road traffic accident. The device is used for implementing the method and has the advantages of simple principle, easy realization, direct reflection of the real driving state of a driver, and improvement of the driving safety.


Patent
National University of Defense Technology | Date: 2014-12-12

The disclosure discloses a dynamic attitude measurement method of a star sensor based on gyros precise angular correlation. On the basis that a dynamic compensation is performed on each of the measurement exposure frames of the star sensor and a fixed star matching vector matrix having dynamic error and noise influence is obtained in a prior art, a transform matrix between every two adjacent measurement frames of the star sensor is precisely measured by a unit including three gyros fixedly coupled with the star sensor. The transform matrix correlates the matched vector matrixes of the adjacent measurement frames of the star sensor. Finally, a correlated measurement equation is established with a series of correlated measurement frames, which is corresponding to processing a series of measurement frames as a single measurement frame.

Loading National University of Defense Technology collaborators
Loading National University of Defense Technology collaborators