检索结果-南通市图书馆

Hyperspectral image super resolution using deep internal and self-supervised learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第1期9卷 128-141页

作者： Zhe liu xian-Hua Han Graduate School of Science and Technology for Innovation Yamaguchi UniversityYamaguchiJapan Graduate School of Artificial Intelligence and Science Rikkyo UniversityTokyoJapan

By automatically learning the priors embedded in images with powerful modelling ca-pabilities,deep learning-based algorithms have recently made considerable progress in reconstructing the high-resolution hyperspectral(HR-HS)*** previously collected large-amount of external data,these methods are intuitively realised under the full supervision of the ground-truth ***,the database construction in merging the low-resolution(LR)HS(LR-HS)and HR multispectral(MS)or RGB image research paradigm,commonly named as HSI SR,requires collecting corresponding training triplets:HR-MS(RGB),LR-HS and HR-HS image simultaneously,and often faces dif-ficulties in *** learned models with the training datasets collected simultaneously under controlled conditions may significantly degrade the HSI super-resolved perfor-mance to the real images captured under diverse *** handle the above-mentioned limitations,the authors propose to leverage the deep internal and self-supervised learning to solve the HSI SR *** authors advocate that it is possible to train a specific CNN model at test time,called as deep internal learning(DIL),by on-line preparing the training triplet samples from the observed LR-HS/HR-MS(or RGB)images and the down-sampled LR-HS ***,the number of the training triplets extracted solely from the transformed data of the observation itself is extremely few particularly for the HSI SR tasks with large spatial upscale factors,which would result in limited reconstruction *** solve this problem,the authors further exploit deep self-supervised learning(DSL)by considering the observations as the unlabelled training ***,the degradation modules inside the network were elaborated to realise the spatial and spectral down-sampling procedures for transforming the generated HR-HS estimation to the high-resolution RGB/LR-HS approximation,and then the reconstruction errors of the observations were formulated for mea

关键词： computer vision deep learning deep neural networks hyperspectral image enhancement

Quantitative Trait Loci Mapping of Maize Yield and Its Components Under Different Water Treatments at Flowering Time

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Integrative Plant Biology 2006年第10期48卷 1233-1243页

作者： Gui-He Lu Ji-Hua Tang Jian-Bing Yan Xi-Qing Ma Jian-Sheng Li Shao-Jiang Chen Jian-Cang Ma zhan-xian liu Li-Zhu E Yi-Rong zhang Jing-Rui Dai The National Maize Improvement Center of China China Agricultural University Beijing 100094 China Industrial Crop Institute Shanxi Academy of Agricultural Sciences Fenyang 032200 Shanxi China zhangye Maize Seed Farm Zhangye 734024 Gansu China

Drought or water stress is a serious agronomic problem resulting in maize （Zea mays L.） yield loss throughout the world. Breeding hybrids with drought tolerance is one important approach for solving this problem. However, lower efficiency and a longer period of breeding hybrids are disadvantages of traditional breeding programs. It is generally recognized that applying molecular marker techniques to traditional breeding programs could improve the efficiency of the breeding of drought-tolerant maize. To provide useful information for use in studies of maize drought tolerance, the mapping and tagging of quantitative trait loci （QTL） for yield and its components were performed in the present study on the basis of the principle of a mixed linear model. Two hundred and twenty-one recombinant inbred lines （RIL） of Yuyu 22 were grown under both well-watered and water-stressed conditions. In the former treatment group, plants were well irrigated, whereas those in the latter treatment group were stressed at flowering time. Ten plants of each genotype were grown in a row that was 3.00 m × 0.67 m （length × width）. The results show that a few of the QTL were the same （one additive QTL for ear length, two additive QTL and one pair of epistatic QTL for kernel number per row, one additive QTL for kernel weight per plant）, whereas most of other QTL were different between the two different water treatment groups. It may be that genetic expression differs under the two different water conditions. Furthermore, differences in the additive and epistatic QTL among the traits under water-stressed conditions indicate that genetic expression also differs from trait to trait. Major and minor QTL were detected for the traits, except for kernel number per row, under water-stressed conditions. Thus, the genetic mechanism of drought tolerance in maize is complex because the additive and epistatic QTL exist at the same time and the major and minor QTL all contribute to phenotype under water-stres

关键词： flowering time maize (Zea mays) quantitative trait loci mapping water treatment yield and its components

Evaluation of transmission risk of respiratory particles under different ventilation strategies in an elevator

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Building Simulation 2024年第5期17卷 771-784页

作者： Liangyu Zhu xian Li Bujin Feng Fan liu School of Civil Engineering and Architecture Linyi UniversityLinyi276000China College of Agriculture Shandong Agricultural UniversityTaian271018China School of Environment and Architecture University of Shanghai for Science and TechnologyShanghai200093China

People in elevators are at risk of respiratory infection because the elevator cabin is crowded and has poor *** exhaled particles may be inhaled by the susceptible person,deposited on the surface and suspended in the elevator,which can result in direct and indirect ***,whether the air vent designs adopted in the elevator can effectively reduce the transmission risk of respiratory particles remains *** this study,the dispersion of particles under four common ventilation strategies used in the commercial elevator was investigated by proven computational fluid dynamics(CFD)*** flow field was simulated with the RNG κ-ξ turbulence model and the Lagrangian method was adopted to track particle *** effects of air vent layout and airflow rate on particle transmission were *** found that more than 50% of exhaled particles(average value)were suspended in the cabin and difficult to discharge under the investigated ventilation *** deposited fraction of particles on the susceptible person reached up to 39.14% for infiltration ventilation,which led to a high risk of contact *** the ventilation rate could not significantly reduce the inhalation proportion of particles due to the poor airflow distribution inside the elevator.A more proper ventilation strategy should be explored for the elevator to control transmission risk.

关键词： particle dispersion transmission risk elevator ventilation strategy computational fluid dynamics(CFD)

Experimental investigation of engineered geopolymer composite for structural strengthening against blast loads

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第2期32卷 496-509页

作者： Shan liu Chunyuan liu Yifei Hao Yi zhang Li Chen zhan Li Key Laboratory of Coast Civil Structure Safety of Ministry of Education Tianjin UniversityTianjin 300072China Tianjin Key Laboratory of Prefabricated Buildings and Intelligent Construction Hebei University of TechnologyTianjin 300400China Institute of Defense Engineering AMSPLABeijing 100850China Engineering Research Center of Safety and Protection of Explosion&Impact of Ministry of Education Southeast UniversityNanjing 211189JiangsuChina State Key Laboratory of Disaster Prevention&Mitigation of Explosion&Impact Army Engineering University of the PLANanjing 210007JiangsuChina

The recent increase in blast/bombing incidents all over the world has pushed the development of effective strengthening approaches to enhance the blast resistance of existing civil *** geopolymer composite(EGC)is a promising material featured by eco-friendly,fast-setting and strain-hardening characteristics for emergent strengthening and ***,the fiber optimization for preparing EGC and its protective effect on structural elements under blast scenarios are *** this study,laboratory tests were firstly conducted to evaluate the effects of fiber types on the properties of EGC in terms of workability,dry shrinkage,and mechanical properties in compression,tension and *** experimental results showed that EGC containing PE fiber exhibited suitable workability,acceptable dry shrinkage and superior mechanical properties compared with other types of *** that,a series of field tests were carried out to evaluate the effectiveness of EGC retrofitting layer on the enhancement of blast performance of typical *** tests include autoclaved aerated concrete(AAC)masonry walls subjected to vented gas explosion,reinforced AAC panels subjected to TNT explosion and plain concrete slabs subjected to contact *** was found that EGC could effectively enhance the blast resistance of structural elements in different *** AAC masonry walls and panels,with the existence of EGC,the integrity of specimens could be maintained,and their deflections and damage were significantly *** plain concrete slabs,the EGC overlay could reduce the diameter and depth of the crater and spallation of specimens.

关键词： Engineered geopolymer composites Fiber optimization Strengthening material Blast resistance Masonry wall Reinforced AAC panel Plain concrete slab

Enhancing action discrimination via category-specific frame clustering for weakly-supervised temporal action localization

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2024年第6期25卷 809-823页

作者： Huifen XIA Yongzhao zhan Honglin liu Xiaopeng REN School of Computer Science and Communication Engineering Jiangsu UniversityZhenjiang 212013China Jiangsu Engineering Research Center of Big Data Ubiquitous Perception and Intelligent Agricultural Applications Zhenjiang 212013China Changzhou Vocational Institute of Mechatronic Technology Changzhou 213164China

Temporal action localization (TAL) is a task of detecting the start and end timestamps of action instances and classifying them in an untrimmed video. As the number of action categories per video increases, existing weakly-supervised TAL (W-TAL) methods with only video-level labels cannot provide sufficient supervision. Single-frame supervision has attracted the interest of researchers. Existing paradigms model single-frame annotations from the perspective of video snippet sequences, neglect action discrimination of annotated frames, and do not pay sufficient attention to their correlations in the same category. Considering a category, the annotated frames exhibit distinctive appearance characteristics or clear action ***, a novel method to enhance action discrimination via category-specific frame clustering for W-TAL is proposed. Specifically,the K-means clustering algorithm is employed to aggregate the annotated discriminative frames of the same category, which are regarded as exemplars to exhibit the characteristics of the action category. Then, the class activation scores are obtained by calculating the similarities between a frame and exemplars of various categories. Category-specific representation modeling can provide complimentary guidance to snippet sequence modeling in the mainline. As a result, a convex combination fusion mechanism is presented for annotated frames and snippet sequences to enhance the consistency properties of action discrimination,which can generate a robust class activation sequence for precise action classification and localization. Due to the supplementary guidance of action discriminative enhancement for video snippet sequences, our method outperforms existing single-frame annotation based methods. Experiments conducted on three datasets (THUMOS14, GTEA, and BEOID) show that our method achieves high localization performance compared with state-of-the-art methods.

关键词： Weakly supervised Temporal action localization Single-frame annotation Category-specific Action discrimination

Experimental Investigation on Vertical Hydraulic Transport of Ores in Deepsea Mining

在线全文

学校读者我要写书评

暂无评论

China Ocean Engineering 2024年第1期38卷 42-53页

作者： liu Lei liu Jian-cheng LI Xin XU Li-xin zhanG Xiu-zhan State Key Laboratory of Ocean Engineering Shanghai Jiao Tong UniversityShanghai 200240China School of Naval Architecture Ocean and Civil EngineeringShanghai Jiao Tong UniversityShanghai 200240China Sanya Yazhou Bay Institution of Deep-sea Sci-Tech Shanghai Jiao Tong UniversitySanya 572000China China Merchants Marine And Offshore Research Institute Co. Ltd.Shenzhen 518067China China Merchants Deepsea Research Institute(Sanya)Co. Ltd.Sanya 572000China School of Naval Architecture and Ocean Engineering Jiangsu University of Science and TechnologyZhenjiang 212100China

Deepsea mining has been proposed since the 1960s to alleviate the lack of resources on *** hydraulic transport of collected ores from the seabed to the sea surface is considered the most promising method for industrial *** the present study,an indoor model test of the vertical hydraulic transport of particles was conducted.A noncontact optical method has been proposed to measure the local characteristics of the particles inside a vertical pipe,including the local concentration and particle *** hydraulic gradient of ore transport was evaluated with various particle size distributions,particle densities,feeding concentrations and mixture flow *** transport,the local concentration is larger than the feeding concentration,whereas the particle velocity is less than the mixture *** qualitative effects of the local concentration and local fluid velocity on the particle velocity and slip velocity were *** local fluid velocity contributes significantly to particle velocity and slip velocity,whereas the effect of the local concentration is marginal.A higher feeding concentration and mixture flow velocity result in an increased hydraulic *** effect of the particle size gradation is slight,whereas the particle density plays a crucial role in the transport.

关键词： deepsea mining vertical hydraulic transport experiment flow characteristics

Iron-nickel alloy particles with N-doped carbon “armor” as a highly selective and long-lasting catalyst for the synthesis of Nbenzylaniline molecules

在线全文

学校读者我要写书评

暂无评论

Nano Research 2024年第4期17卷 2308-2319页

作者： Gang Wang Longchao Sun Wanyi liu Haijuan zhan Shuxian Bi College of Chemistry and Chemical Engineering Ningxia UniversityYinchuan 750021China

A scalable strategy for the convenient and rapid preparation of nitrogen-doped carbon-coated iron-based alloy catalysts was *** controlling the type and amount of metal salts in the precursor,various types of nitrogen-doped carbon-coated alloy catalysts can be prepared in a targeted ***_(2)Ni2@CN materials with small particle sizes and relatively homogeneous basic sites showed promising results in the N-alkylation reaction of benzyl alcohol with aniline(optimum yield:99%).It is worth noting that the catalyst can also be magnetically separated and recovered after the reaction,and its performance can be regenerated through simple ***,it was confirmed by kinetic experiments that the activation of C–H at the benzyl alcohol benzylic position is the rate-determining step(RDS).According to density flooding theory calculations,Fe_(2)Ni2@CN catalysts require less energy than other materials(Fe@CN and Ni@CN)for the RDS(dehydrogenation reaction)*** N-alkylation reactions are more easily carried out on Fe_(2)Ni2@CN catalysts,which may be the reason for the best catalytic activity of Fe-Ni alloy *** carbon-coated alloy materials will show great potential in more types of heterogeneous catalysis.

关键词： iron-nickel alloy catalyst N-alkylation reaction magnetic separation reaction path dehydrogenation reaction

Spontaneous Recovery in Directed Dynamical Networks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Engineering 2024年第6期37卷 208-214页

作者： Xueming liu xian Yan H.Eugene Stanley School of Artificial Intelligence and Automation&The MOE Engineering Research Center of Autonomous Intelligent Unmanned Systems&the Key Laboratory of Image Processing Huazhong University of Science and TechnologyWuhan 430074China Center for Polymer Studies Department of PhysicsBoston UniversityBostonMA 02215USA

Complex networked systems,which range from biological systems in the natural world to infrastructure systems in the human-made world,can exhibit spontaneous recovery after a failure;for example,a brain may spontaneously return to normal after a seizure,and traffic flow can become smooth again after a *** studies on the spontaneous recovery of dynamical networks have been limited to undirected ***,most real-world networks are *** fill this gap,we build a model in which nodes may alternately fail and recover,and we develop a theoretical tool to analyze the recovery properties of directed dynamical *** find that the tool can accurately predict the final fraction of active nodes,and the prediction accuracy decreases as the fraction of bidirectional links in the network increases,which emphasizes the importance of directionality in network *** to different initial states,directed dynamical networks may show alternative stable states under the same control parameter,exhibiting hysteresis *** addition,for networks with finite sizes,the fraction of active nodes may jump back and forth between high and low states,mimicking repetitive failure-recovery *** findings could help clarify the system recovery mechanism and enable better design of networked systems with high resilience.

关键词： Network resilience Directed dynamical networks Spontaneous recovery

Controlled flight of a self-powered micro blimp driven by insect-sized flapping-wing thrusters

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第6期37卷 127-136页

作者： xian YU Zhiwei liu Jiaming LENG Mingjing QI Xiaojun YAN School of Energy and Power Engineering Beihang UniversityBeijing 100191China Collaborative Innovation Center of Advanced Aero-Engine Beijing 100191China National Key Laboratory of Science and Technology on Aero-Engine Aero-thermodynamics Beihang UniversityBeijng 100191China Beijing Key Laboratory of Aero-Engine Structure and Strength Beihang UniversityBeijing 100191China

Micro aerial platforms face significant challenges in achieving long controlled endurance as most of the energy is consumed to overcome the weight of the *** this study,we present a controllable micro blimp that addresses this issue through the use of a helium-filled *** micro blimp has a long axis of 23 cm and is propelled by four insect-sized flapping-wing thrusters,each weighing 80 mg and with a wingspan of 3.5 *** distributed thrusters enable controlled motions and provide the micro blimp with an advantage in flight endurance compared to multirotors or flapping-wing micro aerial vehicles at the same size *** enhance the performance of the controlled flight,we propose a wireless control module that enables manipulation from a distance of up to 100 ***,a smartphone application is developed to send instructions to the circuit board,allowing the blimp to turn left and right,ascend and descend,and achieve a combination of these movements *** findings demonstrate that this micro blimp is one of the smallest controlled self-powered micro blimps to date.

关键词： Microairvehicle Aircraft control Self-powered micro blimp Flapping-wing thruster Wireless flight