检索结果-南通市图书馆

Unsupervised Meteorological Downscaling Based on Dual Learning and Subgrid-scale Auxiliary Information

Advances in Atmospheric Sciences 2025年第1期42卷 53-66页

作者： Jing HU Jialing MU xiaomeng HUANG xi wu Department of Computer Science Chengdu University of Information TechnologyChengdu 610225China Department of Earth System Science Ministry of Education Key Laboratory for Earth System ModelingInstitute for Global Change StudiesTsinghua UniversityBeijing 100084China

Climate downscaling is used to transform large-scale meteorological data into small-scale data with enhanced detail,which finds wide applications in climate modeling,numerical weather forecasting,and renewable *** deeplearning-based downscaling methods effectively capture the complex nonlinear mapping between meteorological data of varying scales,the supervised deep-learning-based downscaling methods suffer from insufficient high-resolution data in practice,and unsupervised methods struggle with accurately inferring small-scale specifics from limited large-scale inputs due to small-scale *** article presents DualDS,a dual-learning framework utilizing a Generative Adversarial Network–based neural network and subgrid-scale auxiliary information for climate *** a learning method is unified in a two-stream framework through up-and downsamplers,where the downsampler is used to simulate the information loss process during the upscaling,and the upsampler is used to reconstruct lost details and correct errors incurred during the *** dual learning strategy can eliminate the dependence on high-resolution ground truth data in the training process and refine the downscaling results by constraining the mapping *** findings demonstrate that DualDS is comparable to several state-of-the-art deep learning downscaling approaches,both qualitatively and ***,for a single surface-temperature data downscaling task,our method is comparable with other unsupervised algorithms with the same dataset,and we can achieve a 0.469 dB higher peak signal-to-noise ratio,0.017 higher structural similarity,0.08 lower RMSE,and the best correlation *** summary,this paper presents a novel approach to addressing small-scale uncertainty issues in unsupervised downscaling processes.

关键词： downscaling unsupervised deep learning dual learning auxiliary information

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Crop root system architecture in drought response

引用

Journal of Genetics and Genomics 2025年第1期52卷 4-13页

作者： Yanjun Zhang xi wu xingrong Wang Mingqiu Dai Yunling Peng College of Agronomy Gansu Agricultural UniversityLanzhouGansu 730070China State Key Laboratory of Aridland Crop Science Gansu Agricultural UniversityLanzhouGansu 730070China Crop Research Institute Gansu Academy of Agricultural SciencesLanzhouGansu 730070China Key Laboratory of Crop Gene Resources and Germplasm Innovation in Northwest Cold and Arid Regions Ministry of Agriculture and Rural AffairsLanzhouGansu 730070China National Key Laboratory of Crop Genetic Improvement Huazhong Agricultural UniversityWuhanHubei 430070China Hubei Hongshan Laboratory WuhanHubei 430070China

Drought is a natural disaster that profoundly impacts on global agricultural production,significantly reduces crop yields,and thereby poses a severe threat to worldwide food *** the challenge of effectively improving crop drought resistance(DR)to mitigate yield loss under drought conditions is a global *** optimal root system architecture(RSA)plays a pivotal role in enhancing the capacity of crops to efficiently uptake water and nutrients,which consequently strengthens their resilience against environmental *** this review,we discuss the compositions and roles of crop RSA and summarize the most recent developments in augmenting drought tolerance in crops by manipulating RSA-related *** on the current research,we propose the potential optimal RSA configuration that could be helpful in enhancing crop ***,we discuss the existing challenges and future directions for breeding crops with enhanced DR capabilities through genetic improvements targeting RSA.

关键词： Crop Drought resistance Root system architecture Genetic improvement

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Physically Constrained Adaptive Deep Learning for Ocean Vertical-Mixing Parameterization

引用

Advances in Atmospheric Sciences 2025年第1期42卷 165-177页

作者： Junjie FANG xiaojie LI Jin LI Zhanao HUANG Yongqiang YU xiaomeng HUANG xi wu Chengdu University of Information Technology Chengdu 610225China Institute of Atmospheric Physics Chinese Academy of SciencesBeijing 100029China Tsinghua University Beijing 100084China

Existing traditional ocean vertical-mixing schemes are empirically developed without a thorough understanding of the physical processes involved,resulting in a discrepancy between the parameterization and forecast *** uncertainty in ocean-mixing parameterization is primarily responsible for the bias in ocean *** from deep-learning technology,we design the Adaptive Fully Connected Module with an Inception module as the baseline to minimize *** adaptively extracts the best features through fully connected layers with different widths,and better learns the nonlinear relationship between input variables and parameterization ***,to obtain more accurate results,we impose KPP(K-Profile Parameterization)and PP(Pacanowski–Philander)schemes as physical constraints to make the network parameterization process follow the basic physical laws more *** model data are calculated with human experience,lacking some unknown physical processes,which may differ from the actual data,we use a decade-long time record of hydrological and turbulence observations in the tropical Pacific Ocean as training *** physical constraints and a nonlinear activation function,our method catches its nonlinear change and better adapts to the oceanmixing parameterization *** use of physical constraints can improve the final results.

关键词： deep learning vertical-mixing parameterization ocean sciences adaptive network

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

受腹侧通路启发的脂肪肝超声图像分类方法VPNet

引用

计算机应用 2025年第2期45卷 662-669页

作者：丁丹妮彭博吴锡成都信息工程大学计算机学院成都610225 西南交通大学计算机与人工智能学院成都611756

考虑腹侧通路在视觉信息处理中的核心作用,提出一种基于腹侧通路的脂肪肝分类方法。通过整合卷积神经网络(CNN)和生物视觉认知模型,模拟从初级视觉皮层(V1)到下颞叶皮层(IT Cortex)的层次化信息加工流程,从而构建全新的神经网络架构——... 详细信息

考虑腹侧通路在视觉信息处理中的核心作用,提出一种基于腹侧通路的脂肪肝分类方法。通过整合卷积神经网络(CNN)和生物视觉认知模型,模拟从初级视觉皮层(V1)到下颞叶皮层(IT Cortex)的层次化信息加工流程,从而构建全新的神经网络架构——VPNet (Ventral Pathway Network)。此外,受生物视觉机制中非经典感受野(nCRF)抑制机制在背景噪声抑制方面的启发,模拟该机制以应对超声图像中斑点噪声的挑战,进而增强模型的特征识别能力。在自制数据集上进行四类别的脂肪肝变异程度识别时,VPNet达到88.37%的准确率;在公开数据集上进行二类别的脂肪肝诊断时,VPNet的准确率、敏感性和特异性均达到100%的最佳性能。实验结果表明,与已知公开数据集研究中较优的ResNet101-SVM相比,VPNet的准确率分别在自制数据集和公开数据集上提升了11.63和0.7个百分点,证明了所提方法在脂肪肝疾病诊断中的有效性。

关键词：超声图像分类脂肪肝变性生物视觉腹侧通路非经典感受野抑制深度学习

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

联合生成对抗网络与扩散模型的稀疏CT图像重建

引用

软件导刊 2025年第2期24卷 172-180页

作者：钟泉吴锡成都信息工程大学计算机学院四川成都610225

稀疏CT图像重建对降低患者辐射剂量并促进影像诊断具有重要的临床意义。在基于深度学习的医学图像重建任务中,现有方法往往忽略了真实图像与重建图像之间的残差,造成重建图像错误结构的生成和细节的不足。生成对抗网络通过对抗学习快速... 详细信息

稀疏CT图像重建对降低患者辐射剂量并促进影像诊断具有重要的临床意义。在基于深度学习的医学图像重建任务中,现有方法往往忽略了真实图像与重建图像之间的残差,造成重建图像错误结构的生成和细节的不足。生成对抗网络通过对抗学习快速重建全局内容和结构信息,而扩散模型通过稳定的训练重建出细节丰富的图像。为提高稀疏CT重建图像质量,提出一种联合生成对抗网络与扩散模型的残差细化重建网络(RRRNet)。该网络首先利用生成对抗网络作为初级生成器,捕捉图像的全局结构信息,然后通过扩散模型建模真实数据与初步预测之间的残差以修正初步预测。此外,在细化过程中还引入高频信息分离训练模块,以增强边缘和细节的恢复能力。在LIDC数据集上的实验结果表明,在4.50%的采样率下,RRRNet在SSIM、PSNR和MAE等量化指标的评估上分别达到了96.40%、40.76dB与32.49HU。相比基于生成对抗网络或扩散模型的方法,RRRNet提升了重建图像质量。

关键词：图像重建深度学习生成对抗网络扩散模型

来源：

维普期刊数据库

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

8-ranks of Class Groups of Some Imaginary Quadratic Number Fields

引用

Acta Mathematica Sinica,English Series 2007年第11期23卷 2061-2068页

作者： xi Mei wu Department of Mathematics Nanjing University of Aeronautics and AstronauticsNanjing 210016 P. R. China State Key Laboratory of Information Security Graduate School of Chinese Academy of Sciences Beijing 100039 P. R. China

Let F = Q（√-p1p2） be an imaginary quadratic field with distinct primes p1 = p2 = 1 mod 8 and the Legendre symbol （p1/p2） = 1. Then the 8-rank of the class group of F is equal to 2 if and only Pl if the following conditions hold：（1） The quartic residue symbols （p1/p2）4 = （p2/p1）4 = 1; （2） Either both p1 and p2 are represented by the form a^2 ＋ 32b^2 over Z and p^h2＋（2p1）/4=x^2-2p1y^2,x,y∈Z,or both p1 and p2 are not represented by the form a^2 ＋ 32b^2 over Z and p^h2＋（2p1）/4=ε（2x^2-p1y^2）,x,y∈Z,ε∈{±1},where h＋（2p1） is the narrow class number of Q（√2p1）,Moreover, we also generalize these results.

关键词： class group Rédei matrix reciprocity law

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Electric conductivity and electric convertibility of potassium acetate in water, ethanol, 2,2,2–trifluoroethanol, 2–propanol and their binary blends

引用

Chinese Journal of Chemical Engineering 2018年第12期26卷 2581-2591页

作者： xi wu Shiming Xu Debing wu Huan Liu Key Laboratory of Ocean Energy Utilization and Energy Conservation of Ministry of Education School of Energy and Power Engineering Dalian University of Technology

Salinity gradient energy between the concentrated and diluted electrolyte solutions can be converted to electric energy by using reverse electrodialysis(RED) technology. Electrolyte solution is a vital factor that impacts the energy conversion efficiency. Potassium acetate(KAc) was chosen as solute, and water, ethanol, 2,2,2-trifluoroethanol(TFE), 2-propanol(IPA) and several of their binary mixtures were selected as solvents. Electric conductivity of these solutions were measured under varying conditions. KAc was easily ionized in water and possessed the maximum electric conductivity, following by KAc–H_2O–TFE and KAc–H_2O–ethanol, and then KAc in pure TFE, ethanol, and IPA respectively. For electric convertibility of these solutions working in a RED power generation system, it was found that the KAc–H_2O possessed the maximum power density, and the KAc–ethanol–H_2O possessed the larger open circuit voltage than aqueous KAc solution under the same working condition. Besides, it was observed that both the electric conductivity and electric convertibility were significantly influenced by the concentration and temperature of solution. With the increasing of concentration,electric conductivity of these solutions increased firstly and then reached to the peak, but later it *** temperature took a positive impact role to the electric conductivity. Electric conductivity of these solutions can be estimated by using a modified amplitude version of Gaussian peak function.

关键词： Conductivity Alcohol Electrochemistry Solution Mixtures

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

肉桂醛及肉桂提取液在结直肠炎症与癌症防治中作用机制的研究进展

引用

山东医药 2025年第1期65卷 119-122,128页

作者：杨宏丽吴熙国家癌症中心/国家肿瘤临床医学研究中心/中国医学科学院北京协和医学院肿瘤医院中医科北京100021 国家癌症中心/国家肿瘤临床医学研究中心/中国医学科学院北京协和医学院肿瘤医院综合内科北京100021

炎症性肠病是一种慢性非特异性肠道炎症性疾病,由于缺乏特异性治疗手段,往往反复发作、迁延难愈,并有可能进一步进展为结直肠癌。肉桂为樟科樟属植物肉桂的干燥树皮,具有补火助阳、引火归元、散寒止痛、温通经脉的功效,主要用于治疗腹... 详细信息

炎症性肠病是一种慢性非特异性肠道炎症性疾病,由于缺乏特异性治疗手段,往往反复发作、迁延难愈,并有可能进一步进展为结直肠癌。肉桂为樟科樟属植物肉桂的干燥树皮,具有补火助阳、引火归元、散寒止痛、温通经脉的功效,主要用于治疗腹泻、腹痛、胸痹等病症。肉桂中含有大量挥发油,其主要活性成分为肉桂醛。现代药理学研究发现,肉桂醛及肉桂提取液在抗炎、抗氧化、免疫调节、抑菌、抗癌等方面均有良好效果,其抗炎、抗氧化和抗癌特性在结直肠癌防治中的作用尤其明显,不仅可直接抑制结直肠癌细胞增殖,还可通过抗炎、抗氧化作用延缓或阻止正常细胞向结直肠癌细胞的恶性转化和进展。因此,肉桂有望成为预防和治疗结直肠炎症与癌症的多靶点新型植物药物。

关键词：肉桂醛肉桂提取液炎症性肠病结直肠癌

来源：

维普期刊数据库

同方期刊数据库

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

Characterization of carboxylesterase PxαE8 and its role in multi-insecticide resistance in Plutella xylostella(L.)

引用

Journal of Integrative Agriculture 2022年第6期21卷 1713-1721页

作者： LI Ran SUN xi LIANG Pei GAO xi-wu Department of Entomology China Agricultural UniversityBeijing 100193P.R.China

Carboxylesterase(CarE)was considered as important phase-I detoxifying enzymes which participated in detoxification of different types of ***-regulation of CarE genes has been proved playing a major role in insecticide resistance in many pest insects,but its involvement in resistance to insecticides in Plutella xylostella has been rarely *** this study,a CarE cDNA named PxαE8 was identified in ***,which has an open reading frame of 1599 nucleotides and putatively encodes 532 amino *** investigation of spatial expression profiles of PxαE8 revealed that it was expressed in all developmental stages,especially in larvae and *** body part/tissue-specific expression profiles showed that the PxαE8 mainly expressed in fat body,malpighian tubule and hemolymph of ***,the relative expression of PxαE8 in two multi-resistant field populations,Hainan(HN)and Guangdong(GD)populations,was found 24.4-and 15.5-fold higher than that in susceptible population,*** of PxαE8 by RNA interference dramatically increased the mortalities of larvae of HN population treated with LC_(50) of beta-cypermethrin and phoxim by 25.3 and 18.3%,*** results suggested that up-regulation of PxαE8 was involved in resistance to both beta-cypermethrin and phoxim in ***,which shed light on further understanding of molecular mechanisms of multi-insecticide-resistance in *** and other pest insects.

关键词： Plutella xylostella carboxylesterase insecticide resistance beta-cypermethrin phoxim

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Length-Changeable Incremental Extreme Learning Machine

引用

Journal of Computer Science & Technology 2017年第3期32卷 630-643页

作者： You-xi wu Dong Liu He Jiang School of Computer Science and Software Hebei University of Technology Tianjin 300130 China School of Economics and Management Hebei University of Technology Tianjin 300130 China Hebei Province Key Laboratory of Big Data Calculation Tianjin 300401 China School of Software Dalian University of Technology Dalian 116621 China

Extreme learning machine （ELM） is a learning algorithm for generalized single-hidden-layer feed-forward networks （SLFNs）. In order to obtain a suitable network architecture, Incremental Extreme Learning Machine （I-ELM） is a sort of ELM constructing SLFNs by adding hidden nodes one by one. Although kinds of I-ELM-class algorithms were proposed to improve the convergence rate or to obtain minimal training error, they do not change the construction way of I-ELM or face the over-fitting risk. Making the testing error converge quickly and stably therefore becomes an important issue. In this paper, we proposed a new incremental ELM which is referred to as Length-Changeable Incremental Extreme Learning Machine （LCI-ELM）. It allows more than one hidden node to be added to the network and the existing network will be regarded as a whole in output weights tuning. The output weights of newly added hidden nodes are determined using a partial error-minimizing method. We prove that an SLFN constructed using LCI-ELM has approximation capability on a universal compact input set as well as on a finite training set. Experimental results demonstrate that LCI-ELM achieves higher convergence rate as well as lower over-fitting risk than some competitive I-ELM-class algorithms.

关键词： single-hidden-layer feed-forward network (SLFN) incremental extreme learning machine (I-ELM) random hidden node convergence rate universal approximation

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：