期刊文献+
共找到56,646篇文章
< 1 2 250 >
每页显示 20 50 100
Evidence and explanation for the involvement of the nucleus accumbens in pain processing 预览
1
作者 Haley N.Harris Yuan B.Peng 《中国神经再生研究:英文版》 SCIE CAS CSCD 2020年第4期597-605,共9页
The nucleus accumbens(NAc)is a subcortical brain structure known primarily for its roles in pleasure,reward,and addiction.Despite less focus on the NAc in pain research,it also plays a large role in the mediation of p... The nucleus accumbens(NAc)is a subcortical brain structure known primarily for its roles in pleasure,reward,and addiction.Despite less focus on the NAc in pain research,it also plays a large role in the mediation of pain and is effective as a source of analgesia.Evidence for this involvement lies in the NAc’s cortical connections,functions,pharmacology,and therapeutic targeting.The NAc projects to and receives information from notable pain structures,such as the prefrontal cortex,anterior cingulate cortex,periaqueductal gray,habenula,thalamus,etc.Additionally,the NAc and other pain-modulating structures share functions involving opioid regulation and motivational and emotional processing,which each work beyond simply the rewarding experience of pain offset.Pharmacologically speaking,the NAc responds heavily to painful stimuli,due to its high density ofμopioid receptors and the activation of several different neurotransmitter systems in the NAc,such as opioids,dopamine,calcitonin gene-related peptide,γ-aminobutyric acid,glutamate,and substance P,each of which have been shown to elicit analgesic effects.In both preclinical and clinical models,deep brain stimulation of the NAc has elicited successful analgesia.The multi-functional NAc is important in motivational behavior,and the motivation for avoiding pain is just as important to survival as the motivation for seeking pleasure.It is possible,then,that the NAc must be involved in both pleasure and pain in order to help determine the motivational salience of positive and negative events. 展开更多
关键词 analgesia CIRCUITRY deep brain stimulation NOCICEPTION nucleus ACCUMBENS PAIN PAIN relief PAIN signaling REWARD STRIATUM
在线阅读 下载PDF
Deep residual learning for denoising Monte Carlo renderings
2
作者 Kin-Ming Wong Tien-Tsin Wong 《计算可视媒体(英文版)》 CSCD 2019年第3期239-255,共17页
Learning-based techniques have recently been shown to be effective for denoising Monte Carlo rendering methods. However, there remains a quality gap to state-of-the-art handcrafted denoisers. In this paper, we propose... Learning-based techniques have recently been shown to be effective for denoising Monte Carlo rendering methods. However, there remains a quality gap to state-of-the-art handcrafted denoisers. In this paper, we propose a deep residual learning based method that outperforms both state-of-the-art handcrafted denoisers and learning-based denoisers.Unlike the indirect nature of existing learning-based methods(which e.g., estimate the parameters and kernel weights of an explicit feature based filter), we directly map the noisy input pixels to the smoothed output. Using this direct mapping formulation, we demonstrate that even a simple-and-standard ResNet and three common auxiliary features(depth, normal,and albedo) are sufficient to achieve high-quality denoising. This minimal requirement on auxiliary data simplifies both training and integration of our method into most production rendering pipelines. We have evaluated our method on unseen images created by a different renderer. Consistently superior quality denoising is obtained in all cases. 展开更多
关键词 Monte Carlo RENDERING DENOISING DEEP LEARNING DEEP RESIDUAL LEARNING filter-free DENOISING
Classification Method of Encrypted Traffic Based on Deep Neural Network 预览
3
作者 Jing Wan Libing Wu +4 位作者 Youhua Xia Jianzong Hu Zhenchang Xia Rui Zhang Min Wang 《国际计算机前沿大会会议论文集》 2019年第2期542-544,共3页
With the widespread use of network traffic encryption technology, the traditional traffic classification method has gradually become invalid, which increases the difficulty of network management and poses a serious th... With the widespread use of network traffic encryption technology, the traditional traffic classification method has gradually become invalid, which increases the difficulty of network management and poses a serious threat to network security. This paper analyzes the traffic encrypted and transmitted by VPN and explores its classification method. By extracting the timing characteristics of the encrypted traffic, the classification model of the deep neural network was used to classify the traffic of seven different categories in the encrypted traffic, and compared with the commonly used naive Bayesian classification algorithm. At the same time, the batch size that affects the training of deep neural network models was studied. Experiments show that the classification ability of encrypted traffic classification model based on deep neural network is much better than the naive Bayesian method. During training, the batch size has different effects on the deep neural network model. When the batch size is 40, the deep neural network model has the best classification ability. 展开更多
关键词 Encrypted TRAFFIC classification DEEP NEURAL networks DEEP LEARNING SSL/TLS
在线阅读 下载PDF
Deep forest
4
作者 Zhi-Hua Zhou Ji Feng 《国家科学评论:英文版》 CSCD 2019年第1期74-86,共13页
Current deep-learning models are mostly built upon neural networks, i.e. multiple layers of parameterized differentiable non-linear modules that can be trained by backpropagation. In this paper, we explore the possibi... Current deep-learning models are mostly built upon neural networks, i.e. multiple layers of parameterized differentiable non-linear modules that can be trained by backpropagation. In this paper, we explore the possibility of building deep models based on non-differentiable modules such as decision trees. After a discussion about the mystery behind deep neural networks, particularly by contrasting them with shallow neural networks and traditional machine-learning techniques such as decision trees and boosting machines,we conjecture that the success of deep neural networks owes much to three characteristics, i.e.layer-by-layer processing, in-model feature transformation and sufficient model complexity. On one hand,our conjecture may offer inspiration for theoretical understanding of deep learning;on the other hand, to verify the conjecture, we propose an approach that generates deep forest holding these characteristics. This is a decision-tree ensemble approach, with fewer hyper-parameters than deep neural networks, and its model complexity can be automatically determined in a data-dependent way. Experiments show that its performance is quite robust to hyper-parameter settings, such that in most cases, even across different data from different domains, it is able to achieve excellent performance by using the same default setting. This study opens the door to deep learning based on non-differentiable modules without gradient-based adjustment, and exhibits the possibility of constructing deep models without backpropagation. 展开更多
关键词 DEEP FOREST DEEP LEARNING MACHINE LEARNING ENSEMBLE methods DECISION trees
Brain Encoding and Decoding in fMRI with Bidirectional Deep Generative Models 预览
5
作者 Changde Du Jinpeng Li +1 位作者 Lijie Huang Huiguang He 《工程(英文)》 2019年第5期948-953,共6页
Brain encoding and decoding via functional magnetic resonance imaging(fMRI)are two important aspects of visual perception neuroscience.Although previous researchers have made significant advances in brain encoding and... Brain encoding and decoding via functional magnetic resonance imaging(fMRI)are two important aspects of visual perception neuroscience.Although previous researchers have made significant advances in brain encoding and decoding models,existing methods still require improvement using advanced machine learning techniques.For example,traditional methods usually build the encoding and decoding models separately,and are prone to overfitting on a small dataset.In fact,effectively unifying the encoding and decoding procedures may allow for more accurate predictions.In this paper,we first review the existing encoding and decoding methods and discuss the potential advantages of a“bidirectional”modeling strategy.Next,we show that there are correspondences between deep neural networks and human visual streams in terms of the architecture and computational rules.Furthermore,deep generative models(e.g.,variational autoencoders(VAEs)and generative adversarial networks(GANs))have produced promising results in studies on brain encoding and decoding.Finally,we propose that the dual learning method,which was originally designed for machine translation tasks,could help to improve the performance of encoding and decoding models by leveraging large-scale unpaired data. 展开更多
关键词 Brain encoding and decoding Functional magnetic resonance imaging Deep neural networks Deep generative models Dual learning
在线阅读 免费下载
平准化成本在核电项目中的应用研究 预览
6
作者 郑保军 李智勇 +2 位作者 陈向婷 张弘 霍建明 《建筑经济》 北大核心 2019年第12期82-87,共6页
基于IAEA和GIF提出的三种计算平准化成本的方法和工具,以国内成熟的压水堆为案例,深入分析计算方法的差异,对比参数输入和指标计算结果,总结各种方法的异同和优缺点。结合国内核电建设现状和经验,给出计算平准化成本需要关注的事项,尤... 基于IAEA和GIF提出的三种计算平准化成本的方法和工具,以国内成熟的压水堆为案例,深入分析计算方法的差异,对比参数输入和指标计算结果,总结各种方法的异同和优缺点。结合国内核电建设现状和经验,给出计算平准化成本需要关注的事项,尤其说明借鉴三种方法时需要关注的输入参数,能够指导相关从业人员根据国内核电现状对计算步骤和参数进行合理设定,起到规范指标计算、使结果更加准确可信的作用。 展开更多
关键词 平准化成本 INPRO G4ECONS DEEP 核能
在线阅读 下载PDF
深部煤层沿空掘巷围岩稳定性控制技术研究与应用 预览
7
作者 尹英文 王晓菡 马丽妲 《山东煤炭科技》 2019年第4期1-3,共3页
本文研究了深部煤层开采沿空掘巷支护技术方案,通过关键层理论对上覆岩层破断后形成的梁结构进行分析,运用薄板理论对关键层进行分析,根据极限平衡区宽度确定沿空掘巷的小煤柱留设宽度,最终利用高强预应力锚杆进行巷道支护。通过观测王... 本文研究了深部煤层开采沿空掘巷支护技术方案,通过关键层理论对上覆岩层破断后形成的梁结构进行分析,运用薄板理论对关键层进行分析,根据极限平衡区宽度确定沿空掘巷的小煤柱留设宽度,最终利用高强预应力锚杆进行巷道支护。通过观测王楼煤矿七采区深部煤层沿空巷道的支护效果,发现巷道围岩稳定性保持良好。 展开更多
关键词 深部 沿空掘巷 小煤柱 巷道支护
在线阅读 下载PDF
Feather evolution: looking up close and through deep time
8
作者 Xing Xu 《科学通报:英文版》 SCIE EI CSCD 2019年第9期563-564,共2页
The last two decades have witnessed significant advances in our understanding of the evolution of major avian characteristics(1–6)In particular, the spectacular discoveries of fossilized feathered dinosaurs from Chin... The last two decades have witnessed significant advances in our understanding of the evolution of major avian characteristics(1–6)In particular, the spectacular discoveries of fossilized feathered dinosaurs from China and elsewhere have provided key insights into feather evolution (1,2)These fossils indicate that many non-avialan dinosaurs were feathered animals just like living birds. 展开更多
关键词 LOOKING up CLOSE and through DEEP TIME
滇东北铅锌矿山1500m以浅地温梯度变化规律 预览
9
作者 蔡增祥 刘尉俊 +2 位作者 秦云东 王家福 姜允清 《中国矿业》 北大核心 2019年第A01期325-327,330共4页
为了研究滇东北某铅锌矿山1500m以浅的地温梯度变化规律,采用浅孔测温法对矿山四个水平的原岩温度进行测定,并根据测定结果进行拟合分析。实测结果表明:由浅及深的地温梯度分别为0.64℃/hm、0.81℃/hm和1.62℃/hm。由于924m处于开拓阶段... 为了研究滇东北某铅锌矿山1500m以浅的地温梯度变化规律,采用浅孔测温法对矿山四个水平的原岩温度进行测定,并根据测定结果进行拟合分析。实测结果表明:由浅及深的地温梯度分别为0.64℃/hm、0.81℃/hm和1.62℃/hm。由于924m处于开拓阶段,更接近原岩温度,且受裂隙承压水的影响最大,因此地温梯度最大。拟合得出矿山地温梯度为1.449℃/hm,原岩温度随采深的增加逐渐变大,该地温梯度可以用来进一步推测一定开采范围内原始状态下的深部原岩温度,研究结果为该地区深部的热害评价提供了基础和依据。 展开更多
关键词 深部 地温梯度 原岩温度 浅孔测温法
在线阅读 下载PDF
Efficient soluble deep blue electroluminescent dianthracenylphenylene emitters with CIE y(y≤0.08) based on triplet-triplet annihilation
10
作者 Ling Peng Jing-Wen Yao +6 位作者 Mei Wang Lin-Ye Wang Xiao-Lan Huang Xin-Feng Wei Dong-Ge Ma Yong Cao Xu-Hui Zhu 《科学通报:英文版》 SCIE EI CSCD 2019年第11期774-781,共8页
It has been challenging to develop deep blue organic molecular fluorescent emitters with CIE y(y≤0.08)based on triplet-triplet annihilation(TTA). Here, we report facilely available dianthracenylphenylenebased emitter... It has been challenging to develop deep blue organic molecular fluorescent emitters with CIE y(y≤0.08)based on triplet-triplet annihilation(TTA). Here, we report facilely available dianthracenylphenylenebased emitters, which have a 3,5-di(4-t-butylphenyl)phenyl moiety at the one end and 4-cyanophenyl or 3-pyridyl at the other end, respectively. Both fluorophores show a high glass transition temperature of over 220℃ with a thermal decomposition temperature of over 430℃ at an initial weight loss of1%. The preliminary characterizations of the organic light-emitting diodes(OLEDs) that utilized these nondoped emitters provided high EQEs of 4.6%à5.9% with CIE coordinates(0.15, 0.07–0.08). The analysis of the EL transient decay revealed that TTA contributed to the observed performance. The results show that the new emitters are attractive as a potential TTA-based host to afford stable deep blue fluorescent OLEDs. 展开更多
关键词 Anthracene DEEP BLUE emission Fluorescence Organic light-emitting diodes Triplet-triplet annihilation
Soil Nitrogen Distribution and Plant Nitrogen Utilization inDirect-Seeded Rice in Response to Deep Placement of BasalFertilizer-Nitrogen 预览
11
作者 WANG Danying YE Chang +4 位作者 XU Chunmei WANG Zaiman CHEN Song CHU Guang ZHANG Xiufu 《水稻科学:英文版》 CSCD 2019年第6期404-415,共12页
Deep placement of controlled-release fertilizer increases nitrogen (N) use efficiency in rice planting but is expensive. Few studies on direct-seeded rice have examined the effects of deep placement of conventional fe... Deep placement of controlled-release fertilizer increases nitrogen (N) use efficiency in rice planting but is expensive. Few studies on direct-seeded rice have examined the effects of deep placement of conventional fertilizer. With prilled urea serving as N fertilizer, a two-year field experiment with two N rates (120 and 195 kg/hm2) and four basal N application treatments (B50, all fertilizer was broadcast with 50% as basal N;D50, D70 and D100 corresponded to 50%, 70% and 100% of N deeply placed as basal N, respectively) were conducted in direct-seeded rice in 2013 and 2014. Soil N distribution and plant N uptake were analyzed. The results showed that deep placement of basal N significantly increased total N concentrations in soil. Significantly greater soil N concentrations were observed in D100 compared with B50 at 0, 6 and 12 cm (lateral distance) from the fertilizer application point both at mid-tillering and heading stages. D100 presented the highest values of dry matter and N accumulation from seeding to mid-tillering stages, but it presented the lowest values from heading to maturity stages and the lowest grain yield for no sufficient N supply at the reproductive stage. The grain yield of D50 was the highest, however, no significant difference was observed in grain yield, N agronomic efficiency or N recovery efficiency between D70 and D50, or between D70 and B50, while D70 was more labor saving than D50 for only one topdressing was applied in D70 compared with twice in other treatments. The above results indicated that 70% of fertilizer-N deeply placed as a basal fertilizer and 30% of fertilizer-N topdressed as a panicle fertilizer constituted an ideal approach for direct-seeded rice. This recommendation was further verified through on-farm demonstration experiments in 2015, in which D70 produced in similar grain yields as B50 did. 展开更多
关键词 direct-seeded RICE NITROGEN FERTILIZER deep PLACEMENT soil NITROGEN DISTRIBUTION NITROGEN utilization
在线阅读 免费下载
MicroRNA expression in the hippocampal CA1 region under deep hypothermic circulatory arrest 预览
12
作者 Xiao-Hua Wang Dong-Xu Yao +7 位作者 Xiu-Shu Luan Yu Wang Hai-Xia Liu Bei Liu Yang Liu Lei Zhao Xun-Ming Ji Tian-Long Wang 《中国神经再生研究:英文版》 SCIE CAS CSCD 2019年第11期2003-2010,共8页
Using deep hypothermic circulatory arrest, thoracic aorta diseases and complex heart diseases can be subjected to corrective procedures. However, mechanisms underlying brain protection during deep hypothermic circulat... Using deep hypothermic circulatory arrest, thoracic aorta diseases and complex heart diseases can be subjected to corrective procedures. However, mechanisms underlying brain protection during deep hypothermic circulatory arrest are unclear. After piglet models underwent 60 minutes of deep hypothermic circulatory arrest at 14°C, expression of microRNAs(miRNAs) was analyzed in the hippocampus by microarray. Subsequently, TargetScan 6.2, RNA22 v2.0, miRWalk 2.0, and miRanda were used to predict potential targets, and gene ontology enrichment analysis was carried out to identify functional pathways involved. Quantitative reverse transcription-polymerase chain reaction was conducted to verify miRNA changes. Deep hypothermic circulatory arrest altered the expression of 35 miRNAs. Twenty-two miRNAs were significantly downregulated and thirteen miRNAs were significantly upregulated in the hippocampus after deep hypothermic circulatory arrest. Six out of eight targets among the differentially expressed miRNAs were enriched for neuronal projection(cyclin dependent kinase, CDK16 and SLC1 A2), central nervous system development(FOXO3, TYRO3, and SLC1 A2), ion transmembrane transporter activity(ATP2 B2 and SLC1 A2), and interleukin-6 receptor binding(IL6 R)– these are the key functional pathways involved in cerebral protection during deep hypothermic circulatory arrest. Quantitative reverse transcription-polymerase chain reaction confirmed the results of microarray analysis. Our experimental results illustrate a new role for transcriptional regulation in deep hypothermic circulatory arrest, and provide significant insight for the development of miRNAs to treat brain injuries. All procedures were approved by the Animal Care Committee of Xuanwu Hospital, Capital Medical University, China on March 1, 2017(approval No. XW-INI-AD2017-0112). 展开更多
关键词 nerve REGENERATION cerebral protection deep hypothermic circulatory ARREST gene ontology enrichment analysis microRNA hippocampus POST-TRANSCRIPTIONAL expression MICROARRAY BIOINFORMATICS neural REGENERATION
在线阅读 下载PDF
Seizing Construction Opportunity of the Greater Bay Area, Guangdong Shipbuilding and Marine Equipment Industries Riding on Momentum
13
作者 Chen Jianan Kuang Zhanting 《船舶经济贸易》 2019年第4期63-67,共5页
In recent years. China has accelerating advancing into "deep blue". A new bright spot of Guangdong-Hong Kong-Mucao Greater Bay Area's development plan is to cultivate a new growth point of marine economy... In recent years. China has accelerating advancing into "deep blue". A new bright spot of Guangdong-Hong Kong-Mucao Greater Bay Area's development plan is to cultivate a new growth point of marine economy, and cultivating a modern marine industry is an important starting point for the development of marine economy. 展开更多
关键词 accelerating advancing deep blue marine economy
Deep Learning Deepens the Analysis of Alternative Splicing
14
作者 Xudong Zou Xin Gao Wei Chen 《基因组蛋白质组与生物信息学报:英文版》 CAS CSCD 2019年第2期219-221,共3页
The ever-increasing high-volume and high-dimensional geno-mics data on the one hand challenge traditional data analysis approaches,and on the other hand provide ample opportuni-ties for developing novel analytic strat... The ever-increasing high-volume and high-dimensional geno-mics data on the one hand challenge traditional data analysis approaches,and on the other hand provide ample opportuni-ties for developing novel analytic strategies. 展开更多
关键词 DEEP LEARNING Deepens the ANALYSIS ALTERNATIVE SPLICING
Artificial Intelligence Driven Resiliency with Machine Learning and Deep Learning Components 预览
15
作者 Bahman Zohuri Farhang Mossavar Rahmani 《通讯和计算机:中英文版》 2019年第1期1-13,共13页
The future of any business from banking,e-commerce,real estate,homeland security,healthcare,marketing,the stock market,manufacturing,education,retail to government organizations depends on the data and analytics capab... The future of any business from banking,e-commerce,real estate,homeland security,healthcare,marketing,the stock market,manufacturing,education,retail to government organizations depends on the data and analytics capabilities that are built and scaled.The speed of change in technology in recent years has been a real challenge for all businesses.To manage that,a significant number of organizations are exploring the Big Data(BD)infrastructure that helps them to take advantage of new opportunities while saving costs.Timely transformation of information is also critical for the survivability of an organization.Having the right information at the right time will enhance not only the knowledge of stakeholders within an organization but also providing them with a tool to make the right decision at the right moment.It is no longer enough to rely on a sampling of information about the organizations'customers.The decision-makers need to get vital insights into the customers'actual behavior,which requires enormous volumes of data to be processed.We believe that Big Data infrastructure is the key to successful Artificial Intelligence(AI)deployments and accurate,unbiased real-time insights.Big data solutions have a direct impact and changing the way the organization needs to work with help from AI and its components ML and DL.In this article,we discuss these topics. 展开更多
关键词 Artificial INTELLIGENCE RESILIENCE system machine LEARNING DEEP LEARNING BIG data.
在线阅读 下载PDF
Can emergency physicians perform extended compression ultrasound for the diagnosis of lower extremity deep vein thrombosis? 预览
16
作者 Elaine Situ-LaCasse Helpees Guirguis +3 位作者 Lucas Friedman Asad E.Patanwala Seth E.Cohen Srikar Adhikari 《世界急诊医学杂志(英文)》 CAS CSCD 2019年第4期205-209,共5页
BACKGROUND:Current point-of-care ultrasound protocols in the evaluation of lower extremity deep vein thrombosis (DVT) can miss isolated femoral vein clots.Extended compression ultrasound (ECUS) includes evaluation of ... BACKGROUND:Current point-of-care ultrasound protocols in the evaluation of lower extremity deep vein thrombosis (DVT) can miss isolated femoral vein clots.Extended compression ultrasound (ECUS) includes evaluation of the femoral vein from the femoral vein/deep femoral vein bifurcation to the adductor canal.Our objective is to determine if emergency physicians (EPs) can learn ECUS for lower extremity DVT evaluation after a focused training session.METHODS:Prospective study at an urban academic center.Participants with varied ultrasound experience received instruction in ECUS prior to evaluation.Two live models with varied levels of difficult sonographic anatomy were intentionally chosen for the evaluation.Each participant scanned both models.Pre-and post-study surveys were completed.RESULTS:A total of 96 ultrasound examinations were performed by 48 participants (11 attendings and 37 residents).Participants' assessment scores averaged 95.8% (95% CI 93.3%-98.3%) on the easier anatomy live model and averaged 92.3% (95% CI 88.4%-96.2%) on the difficult anatomy model.There were no statistically significant differences between attendings and residents.On the model with easier anatomy,all but 1 participant identified and compressed the proximal femoral vein successfully,and all participants identified and compressed the mid and distal femoral vein.With the difficult anatomy,97.9% (95% CI93.8%-102%) identified and compressed the proximal femoral vein,whereas 93.8% (95% CI 86.9%-100.6%) identified and compressed the mid femoral vein,and 91.7% (95% CI83.9%-99.5%) identified and compressed the distal femoral vein.CONCLUSION:EPs at our institution were able to perform ECUS with good reproducibility after a focused training session. 展开更多
关键词 EMERGENCY medicine POINT-OF-CARE ULTRASOUND Deep VEIN THROMBOSIS
在线阅读 免费下载
A Flower Image Classification Algorithm Based on Saliency Map and PCANet 预览
17
作者 Yan Yangyang Fu Xiang 《通讯和计算机:中英文版》 2019年第1期14-24,共11页
Flower Image Classification is a Fine-Grained Classification problem.The main difficulty of Fine-Grained Classification is the large inter-class similarity and the inner-class difference.In this paper,we propose a new... Flower Image Classification is a Fine-Grained Classification problem.The main difficulty of Fine-Grained Classification is the large inter-class similarity and the inner-class difference.In this paper,we propose a new algorithm based on Saliency Map and PCANet to overcome the difficulty.This algorithm mainly consists of two parts:flower region selection,flower feature learning.In first part,we combine saliency map with gray-scale map to select flower region.In second part,we use the flower region as input to train the PCANet which is a simple deep learning network for learning flower feature automatically,then a 102-way softmax layer that follow the PCANet achieve classification.Our approach achieves 84.12%accuracy on Oxford 17 Flowers dataset.The results show that a combination of Saliency Map and simple deep learning network PCANet can applies to flower image classification problem. 展开更多
关键词 SALIENCY MAP PCANet DEEP LEARNING FLOWER IMAGE classification
在线阅读 下载PDF
Cloud Det ection Using Super Pixel Classification and Semantic Segmentation
18
作者 Han Liu Hang Du +1 位作者 Dan Zeng Qi Tian 《计算机科学技术学报:英文版》 SCIE EI CSCD 2019年第3期622-633,共12页
Cloud detection plays a very significant role in remote sensing image processing.This paper introduces a cloud detection method based on super pixel level classification and semantic segmentation.Firstly,remote sensin... Cloud detection plays a very significant role in remote sensing image processing.This paper introduces a cloud detection method based on super pixel level classification and semantic segmentation.Firstly,remote sensing images are segmented into super pixels.Segmented super pixels compose a super pixel level remote sensing image database.Though cloud detection is essentially a binary classification task,our database is labeled into four categories to improve the generalization ability:thick cloud,cirrus cloud,building,and other culture.Secondly,the super pixel level database is used to train our cloud detection models based on convolution neural network(CNN)and deep forest.Hierarchical fusion CNN is proposed considering super pixel level images contain less semantic information than normal images.Taking full advantage of low-level features like color and texture information,it is more applicable for super pixel level classification.Besides,a distance metric is proposed to refine ambiguous super pixels.Thirdly,an end-to-end cloud detection model based on semantic segmentation is introduced.This model has no restrictions on the input size,and takes less time.Experimental results show that compared with other cloud detection inethods.our proj)osed method acliieves better performance. 展开更多
关键词 CLOUD detection CONVOLUTION NEURAL network deep FOREST SEMANTIC SEGMENTATION
Complexity at Mesoscales:A Common Challenge in Developing Artificial Intelligence 预览
19
作者 Li Guo Jun Wu Jinghai Li 《工程(英文)》 2019年第5期924-929,共6页
Exploring the physical mechanisms of complex systems and making effective use of them are the keys to dealing with the complexity of the world.The emergence of big data and the enhancement of computing power,in conjun... Exploring the physical mechanisms of complex systems and making effective use of them are the keys to dealing with the complexity of the world.The emergence of big data and the enhancement of computing power,in conjunction with the improvement of optimization algorithms,are leading to the development of artificial intelligence(AI)driven by deep learning.However,deep learning fails to reveal the underlying logic and physical connotations of the problems being solved.Mesoscience provides a concept to understand the mechanism of the spatiotemporal multiscale structure of complex systems,and its capability for analyzing complex problems has been validated in different fields.This paper proposes a research paradigm for AI,which introduces the analytical principles of mesoscience into the design of deep learning models.This is done to address the fundamental problem of deep learning models detaching the physical prototype from the problem being solved;the purpose is to promote the sustainable development of AI. 展开更多
关键词 Artificial intelligence Deep learning Mesoscience MESOSCALE Complex system
在线阅读 免费下载
Assessment of strain bursting in deep tunnelling by using the finite-discrete element method 预览
20
作者 Ioannis Vazaios Mark S.Diederichs Nicholas Vlachopoulos 《岩石力学与岩土工程学报:英文版》 CSCD 2019年第1期12-37,共26页
Rockbursting in deep tunnelling is a complex phenomenon posing significant challenges both at the design and construction stages of an underground excavation within hard rock masses and under high in situ stresses.Whi... Rockbursting in deep tunnelling is a complex phenomenon posing significant challenges both at the design and construction stages of an underground excavation within hard rock masses and under high in situ stresses.While local experience,field monitoring,and informed data-rich analysis are some of the tools commonly used to manage the hazards and the associated risks,advanced numerical techniques based on discontinuum modelling have also shown potential in assisting in the assessment of rockbursting.In this study,the hybrid finite-discrete element method(FDEM)is employed to investigate the failure and fracturing processes,and the mechanisms of energy storage and rapid release resulting in bursting,as well as to assess its utility as part of the design process of underground excavations.Following the calibration of the numerical model to simulate a deep excavation in a hard,massive rock mass,discrete fracture network(DFN)geometries are integrated into the model in order to examine the impact of rock structure on rockbursting under high in situ stresses.The obtained analysis results not only highlight the importance of explicitly simulating pre-existing joints within the model,as they affect the mobilised failure mechanisms and the intensity of strain bursting phenomena,but also show how the employed joint network geometry,the field stress conditions,and their interaction influence the extent and depth of the excavation induced damage.Furthermore,a rigorous analysis of the mass and velocity of the ejected rock blocks and comparison of the obtained data with well-established semi-empirical approaches demonstrate the potential of the method to provide realistic estimates of the kinetic energy released during bursting for determining the energy support demand. 展开更多
关键词 ROCKBURST Finite-discrete element method(FDEM) Deep TUNNELLING Hard rock EXCAVATIONS Brittle FRACTURING DISCRETE fracture network(DFN)
在线阅读 下载PDF
上一页 1 2 250 下一页 到第
使用帮助 返回顶部 意见反馈
新型冠状病毒肺炎防控与诊疗专栏