|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Accelerating DNN-based 3D point cloud processing for mobile computing显示文摘3D point cloud data,which are produced by various 3D sensors such as LIDAR and stereo cameras,have been widely deployed by industry leaders such as Google,Uber,Tesla,and Mobileye,for mobile robotic applications such as autonomous driving and humanoid robots.Point cloud data,which are composed of reliable depth information,can provide accurate location and shape characteristics for scene understanding,such as object recognition and semantic segmentation.However,deep neural networks(DNNs),which directly consume point cloud data,are particularly computation-intensive because they have to not only perform multiplication-and-accumulation(MAC)operations but also search neighbors from the irregular 3D point cloud data.Such a task goes beyond the capabilities of general-purpose processors in realtime to figure out the solution as the scales of both point cloud data and DNNs increase from application to application.We present the first accelerator architecture that dynamically configures the hardware onthe-fly to match the computation of both neighbor point search and MAC computation for point-based DNNs.To facilitate the process of neighbor point search and reduce the computation costs,a grid-based algorithm is introduced to search neighbor points from a local region of grids.Evaluation results based on the scene recognition and segmentation tasks show that the proposed design harvests 16.4x higher performance and saves 99.95%of energy than an NVIDIA Tesla K40 GPU baseline in point cloud scene understanding applications. | Bosheng LIU Xiaoming CHEN Yinhe HAN Jiajun LI Haobo XU Xiaowei LI | 2019 | Science China(Information Sciences)2019,62,11: | 4 |
| 2 | Effect of temperature on Lu’an bituminous char structure evolution in pyrolysis and combustion显示文摘In the process of pyrolysis and combustion of coal particles, coal structure evolution will be affected by the ash behavior, which will further affect the char reactivity, especially in the ash melting temperature zone. Lu’an bituminous char and ash samples were prepared at the N_(2) and air atmospheres respectively across ash melting temperature. A scanning electron microscope (SEM) was used to observe the morphology of char and ash. The specific surface area (SSA) analyzer and thermogravimetric analyzer were respectively adopted to obtain the pore structure characteristics of the coal chars and combustion parameters. Besides, an X-ray diffractometer (XRD) was applied to investigate the graphitization degree of coal chars prepared at different pyrolysis temperatures. The SEM results indicated that the number density and physical dimension of ash spheres exuded from the char particles both gradually increased with the increasing temperature, thus the coalescence of ash spheres could be observed obviously above 1100℃. Some flocculent materials appeared on the surface of the char particles at 1300℃, and it could be speculated that β-Si_(3)N_(4) was generated in the pyrolysis process under N_(2). The SSA of the chars decreased with the increasing pyrolysis temperature. Inside the char particles, the micropore area and its proportion in the SSA also declined as the pyrolysis temperature increased. Furthermore, the constantly increasing pyrolysis temperature also caused the reactivity of char decrease, which is consistent with the results obtained by XRD. The higher combustion temperature resulted in the lower porosity and more fragments of the ash. | Yandi ZHANG Yinhe LIU Xiaoli DUAN Yao ZHOU Xiaoqian LIU Shijin XU | 2021 | Frontiers in Energy2021,15,1: | 2 |
| 3 | Analysis of a smoothing method for symmetric conic linear programming显示文摘 | Liu Yongjin Zhang Liwei Wang Yinhe | 2006 | Journal of Applied Mathematics and Computing2006,22,12: | 1 |
| 4 | Pyrolysis and Combustion Characteristics of Coals in Oxyfuel Combustion 显示文摘 | WANG Chang'an ZHANG Xiaoming LIU Yinhe | 2012 | Applied Energy2012,97,26: | 1 |
| 5 | Paleo-Environment Reconstruction of the Middle Permian Lucaogou Formation,Southeastern Junggar Basin,NW China:Implications for the Mechanism of Organic Matter Enrichment in Ancient Lake显示文摘The Middle Permian Lucaogou Formation is the source rock and the main oil shale producing formation in the southeastern Junggar Basin.This study focused on the Lucaogou Formation exposed in two outcrop sections on the northern flank of the Bogda Mountain,namely the Jingjingzigou and Dalongkou sections.Here,we present integrated analysis of the sedimentology,major and trace elements,mineral components and total organic carbon contents.The paleo-environment was reconstructed including provenance,redox conditions,paleo-salinity,chemical weathering intensity and primary organic matter productivity.The results showed that the upper and lower units were deposited in distinct depositional environments with different organic matter accumulation mechanisms.The lower unit was characterized by low lake level,dry climate,fresh-brackish and well-oxygenated water.While during the deposition of the upper unit the lake level rose,climate turned wetter and the bottom water became less oxidized and much saltier.The mechanism of the organic matter accumulation is different for these two units.The preserved organic matters were mainly controlled by the primary productivity in the lower unit and by the redox conditions in the upper unit. | Dawei Cheng Chuanmin Zhou Zhijie Zhang Xuanjun Yuan Yinhe Liu Xingyu Chen | 2022 | Journal of Earth Science2022,33,4: | 1 |
| 6 | Effects of air staging conditions on the combustion and NOx emission characteristics in a600MW wall fired utility boiler using lean coal显示文摘 | LIU Hu LIU Yinhe YI Guangzhou | 2013 | Energy Fuels2013,27,58: | 1 |
| 7 | Pyrolysis and combustion characteristics of coals in oxyfuel combustion显示文摘 | Chang’an Wang Xiaoming Zhang Yinhe Liu Defu Che | 2012 | Applied Energy2012,,: | 1 |
| 8 | Effect of the cooling and reheating during coal pyrolysis on the conversion from char-N to NO/N2O显示文摘 | Jia Qiong Che Defu Liu Yinhe | 2009 | Fuel Processing Technology2009,90,: | 1 |
| 9 | Extracting transient Rayleigh wave and its application in detecting quality of highway roadbed显示文摘 | Liu Jiangping Xia Jianghai Luo Yinhe | 2004 | Progress in Environmental and Engineering Geophysics2004,,: | 1 |
| 10 | Joint Inversion of High-frequency Surface Waves with Fundamental and Higher Modes显示文摘 | Luo Yinhe Xia Jianghai Liu Jiangping | | 0,,: | 1 |
| 11 | Joint inversion of high--frequency surface waves with fundamental and higher modes 显示文摘 | Yinhe Luo Jianghai Xia Jiangping Liu | 2007 | Journal of applied Geophysics2007,62,4: | 1 |
| 12 | Design and simulation of reverse-blocking Schottky-drain AlN/AlGaN HEMTs with drain field plate显示文摘In this paper,Schottky-drain reverse-blocking Al N/Al Ga N HEMTs with drain field plate(FP)have been investigated by Silvaco-ATLAS tools.For HEMTs without FP,with the increase of Al mole fraction in Al Ga N channel from 0 to 0.5,the reverse-blocking voltage increases from-158 V to-720 V.By using the drain field plate technique,a second electric field peak is introduced and the reverse-blocking voltage can be improved.Combined with the optimization of the Si N passivation thickness,optimal electric field management can be achieved to obtain the highest reverse-blocking voltage devices.Since HEMTs with different Al mole fractions possess different critical electric field values,the optimal Si N thickness are varied.With the increase of the Al mole fraction from 0 to 0.5,the reverse-blocking voltage increases from-510 V to-4500 V for HEMTs using drain FP and optimal Si N passivation thickness,and a high power figure-of-merit of 1.171 GW/cm^(2)is achieved.Al Ga N channel HEMTs with Al mole fraction demonstrate great potential for power applications. | Dujun ZHAO Zhaoxi WU Chao DUAN Bo MEI Zhongyang LI Zhongxu WANG Qing TANG Qing YANG Yinhe WU Weihang ZHANG Zhihong LIU Shenglei ZHAO Jincheng ZHANG Yue HAO | 2022 | Science China(Information Sciences)2022,65,2: | 1 |
| 13 | Pulverized coal stream ignition delay under conventional and oxy-fuel combustion conditions显示文摘 | Liu Yinhe Geier Manfred Molina Alejandro | 2011 | International Journal of Greenhouse Gas Control2011,5,1: | 1 |
| 14 | Homogeneous combustion of fuel ultra-lean methane-air mixtures: eeperimental study and simplified reaction mechanism 显示文摘 | Wang Yikun Liu Yinhe Cao Qi | 2011 | Energy & Fuels2011,25,8: | 1 |
| 15 | Releases of NO and its precursors from coal combustion in a fixed bed 显示文摘 | Liu Yinhe Che Defu | 2006 | Fuel Processing Technology2006,87,4: | 1 |
| 16 | Effects of Air Staging Conditions on the Combustion and NO显示文摘 | LIU Hu LIU Yinhe YI Guangzhou | 2013 | Energy and Fu- els2013,27,10: | 1 |
| 17 | Adaptive Control of the Structural Balance for a Class of Complex Dynamical Networks显示文摘In social networks,the structural balance is a state of a group of individuals(nodes)with established mutual relationships(connection relationships)between them.It is easy to see that a social network can be described by a complex dynamical network model composed of the nodes subsystem(NS)and the connection relationships subsystem(CS),where the two subsystems are usually coupled with each other.It implies that the dynamic changes of nodes'states may cause the structural balance in CS.However,few papers have discussed the relationship between the structural balance and the specific dynamic changes of the nodes'states.This paper proposes a model of complex dynamical networks,and mainly focuses on the dynamic changes of states in NS which can lead to the structural balance in CS.It is proved that if each state in NS is doing a specific dynamic motion via the controller with the parameter adaptive law,then the CS can track a given structural balance matrix via the effective coupling and the structural balance can be achieved.Such a result can be regarded as an explanation of the relationship between the structural balance and the specific dynamic changes of the nodes'states.Finally,the simulations verify the effectiveness of the proposed method. | GAO Zilin WANG Yinhe PENG Yi LIU Lizhi CHEN Haoguang | 2020 | Journal of Systems Science & Complexity2020,33,3: | 0 |
| 18 | Application of Multifocusing Method for an Irregular Topography Imaging显示文摘A stack of records becomes one of the main steps in modern seismic data processing. In the stack procedure, the crucial operation is time correction. Conventional methods, e.g., normal moveout (NMO) and dip moveout (DMO) stacks require a sufficiently accurate macro-velocity model, whereas a multifocusing imaging method does not depend on a macro-velocity model. The multifocusing method proposed by Gelchinsky et al. belongs to a group of methods that can be characterized as macro-model-independent imaging methods. The multifocusing method represents a transformation of 2-D multicoverage reflection data into a simulated zero-offset stack profile. This transformation is based on a completely data-derived spatial stacking operator, and includes stacking large supergathers of seismic traces, each of which can span many CMP gathers. By extending the multifocusing moveout formula to explicitly account for non-zero elevations of the source and receiver, the multifocusing imaging method can yield appropriate results when seismic data are acquired over an irregular topography. In recent years, many applications of multifocusing imaging over an irregular topography have demonstrated its advantages in comparison with conventional CMP processing. This paper illustrates the corresponding formulas for a synthetic data example modeled by the wave equation finite difference method. The result of the synthetic example is very encouraging. By stacking large supergathers and applying multifocusing moveout correction, the reflectors are aligned very well and the S/N is greatly improved. We have also applied multifocusing imaging over an irregular topography to a real data example. The elevation of the data acquisition area varies considerably. Applying multifocusing imaging, a substantial improvement of the simulated section was achieved, compared with a conventional CMP stacked section. | Luo Yinhe Liu Jiangping Yao Yao | 2005 | Journal of China University of Geosciences2005,16,3: | 0 |
| 19 | My Views on Gender Issues in China显示文摘I once believed that gender discrimination was not a major issue in China, compared with many other countries. With so many other problems demanding urgent solutions (e.g. poverty, environmental pollution), working on gender issues was like 'gilding the lily'—it was simply a luxury for China at the present stage. My failure to realize the seriousness of the gender problem China derived from at least two aspects. First, Chinese traditional culture emphasizes balance of | LI YINHE LIRED LIU | 1998 | Women of China1998,,11: | 0 |
| 20 | Multi-temporal urban semantic understanding based on GF-2 remote sensing imagery:from tri-temporal datasets to multi-task mapping显示文摘High resolution satellite images are becoming increasingly available for urban multi-temporal semantic understanding.However,few datasets can be used for land-use/land-cover(LULC)classification,binary change detection(BCD)and semantic change detection(SCD)simultaneously because classification datasets always have one time phase and BCD datasets focus only on the changed location,ignoring the changed classes.Public SCD datasets are rare but much needed.To solve the above problems,a tri-temporal SCD dataset made up of Gaofen-2(GF-2)remote sensing imagery(with 11 LULC classes and 60 change directions)was built in this study,namely,the Wuhan Urban Semantic Understanding(WUSU)dataset.Popular deep learning based methods for LULC classification,BCD and SCD are tested to verify the reliability of WUSU.A Siamese-based multi-task joint framework with a multi-task joint loss(MJ loss)named ChangeMJ is proposed to restore the object boundaries and obtains the best results in LULC classification,BCD and SCD,compared to the state-of-the-art(SOTA)methods.Finally,a large spatial-scale mapping for Wuhan central urban area is carried out to verify that the WUsU dataset and the ChangeMJ framework have good application values. | Sunan Shi Yanfei Zhong Yinhe Liu Jue Wang Yuting Wan Ji Zhao Pengyuan Lv Liangpei Zhang Deren Li | 2023 | International Journal of Digital Earth2023,16,1: | 0 |