首页 | 本学科首页   官方微博 | 高级检索  
文章检索
  按 检索   检索词:      
出版年份:   被引次数:   他引次数: 提示:输入*表示无穷大
  收费全文   2753篇
  免费   300篇
  国内免费   350篇
测绘学   1285篇
大气科学   252篇
地球物理   508篇
地质学   544篇
海洋学   327篇
天文学   82篇
综合类   279篇
自然地理   126篇
  2024年   6篇
  2023年   31篇
  2022年   116篇
  2021年   146篇
  2020年   148篇
  2019年   184篇
  2018年   115篇
  2017年   186篇
  2016年   135篇
  2015年   166篇
  2014年   185篇
  2013年   223篇
  2012年   237篇
  2011年   195篇
  2010年   103篇
  2009年   151篇
  2008年   139篇
  2007年   165篇
  2006年   129篇
  2005年   122篇
  2004年   98篇
  2003年   103篇
  2002年   49篇
  2001年   46篇
  2000年   37篇
  1999年   35篇
  1998年   21篇
  1997年   35篇
  1996年   18篇
  1995年   20篇
  1994年   11篇
  1993年   13篇
  1992年   4篇
  1991年   9篇
  1990年   4篇
  1989年   5篇
  1988年   4篇
  1987年   3篇
  1984年   1篇
  1980年   1篇
  1979年   1篇
  1976年   1篇
  1954年   2篇
排序方式: 共有3403条查询结果,搜索用时 31 毫秒
101.
为有效解决传统遥感图像变化检测预处理复杂的问题,提出一种基于随机补片和DeepLabV3+的建筑物遥感图像变化检测方法。以ResNet50特征提取网络为基础,创建DeepLabV3+语义分割网络,并在图像和标签中创建大小为224像素×224像素的随机补片作为网络输入,训练建筑物提取网络;修改建筑物提取网络输入层为6通道,通过矩阵运算将两期遥感图像转换为一幅6通道非RGB图像,利用转换后的非RGB图像进行网络训练并验证变化检测精度。实验1利用ENVI5.3软件,采用马氏距离法进行变化检测;实验2采用改进的U-Net网络和随机补片,完成网络训练和精度验证;实验3使用实验2的训练数据和验证数据,采用随机补片和DeepLabV3+网络进行变化检测网络训练及精度验证。实验结果表明,该方法实验1、实验2、实验3建筑物变化检测平均交并比分别为24.43%、83.14%、89.90%,边界轮廓匹配分数分别为61.47%,80.24%、96.51%。  相似文献   
102.
贾慧涛  刘杨  盛勇 《江苏地质》2021,45(3):335-340
堤坝是重要的水利工程,其安全性至关重要。坝体渗漏是常见的安全隐患之一,对坝体渗漏位置、程度以及坝体结构变化的探测一直是重要的物探课题。微动勘探作为一种新兴的物探手段,应用范围正逐步拓展,通过某水库大坝渗漏检测实例,展示频率-波数法微动线性点阵剖面测量进行大坝渗漏检测的效果,该技术通过提取频散曲线及视S波速度成像,刻画地层速度结构并圈定异常,可实现对大坝渗漏快速精确的无损探测,具有一定的方法优势和广泛的应用前景。  相似文献   
103.
人工智能在冰雹识别及临近预报中的初步应用   总被引:1,自引:0,他引:1  
张文海  李磊 《气象学报》2019,77(2):282-291
基于广东10部S波段多普勒天气雷达的三维拼图资料,利用机器学习技术开发了一种冰雹识别和临近预报的人工智能算法。算法设计时以雷达回波反射率的垂直和水平扫描数据为基础训练集,将冰雹云的雷达反射率扫描数据作为正样本,将其他雷达反射率扫描数据作为负样本,通过贝叶斯分类法对正、负样本数据集进行机器学习,训练人工智能识别冰雹云内在规律的能力。训练时以广东省2008-2013和2015-2016年的数据作为训练集,使用了2014年广东省12次冰雹过程的数据做检验。对比检验的结果表明,人工智能法比传统的概念模型法击中率高9个百分点。研究结果表明了人工智能对冰雹这类非线性强天气过程具有较强的识别能力。   相似文献   
104.
对说话人意图的识别极大地推进了自然语言理解任务的发展.之前的工作大多采用Bi-LSTM即双向LSTM模型进行词汇特征与词汇之间语义关系的提取,但这并不能很好地使句子整体和构成句子的词汇个体之间的信息进行交流.而S-LSTM(Sentence-state LSTM)模型,即句子状态LSTM模型可以很好地将自然语言中句子整体与词汇个体的信息相结合,以便于我们挖掘与利用意图检测与槽值填充之间的关系成立联合模型来更好地理解应答系统中蕴含的语义.因此,本文引入了‘槽值门’机制解决S-LSTM应用于意图检测与槽填充的联合任务时最新迭代时刻的句子状态信息没有得到充分利用的问题.最终的实验结果在ATIS数据集和Snips数据集上均取得了优于目前最先进算法的结果.  相似文献   
105.
Abstract

Variability of river flow is investigated in 502 river flow gauging stations in nine countries of the southern African region with a view to document the spatial variability of the river flow regimes. Those regions where there is strong evidence of declining or increasing trend in annual runoff have been identified. The study has shown that runoff in the region ranges from over 320 mm year?1 in the Lower Zambezi and the highlands of Tanzania to less than 10 mm year?1 in the deserts of Namibia and the Kalahari. There is also evidence of declining runoff in parts of Zambia, Angola, Mozambique and the High Veld in South Africa. The recent decline seems to have started from around 1975.  相似文献   
106.
Abstract

Results of a study on change detection in hydrological time series of annual maximum river flow are presented. Out of more than a thousand long time series made available by the Global Runoff Data Centre (GRDC) in Koblenz, Germany, a worldwide data set consisting of 195 long series of daily mean flow records was selected, based on such criteria as length of series, currency, lack of gaps and missing values, adequate geographical distribution, and priority to smaller catchments. The analysis of annual maximum flows does not support the hypothesis of ubiquitous growth of high flows. Although 27 cases of strong, statistically significant increase were identified by the Mann-Kendall test, there are 31 decreases as well, and most (137) time series do not show any significant changes (at the 10% level). Caution is advised in interpreting these results as flooding is a complex phenomenon, caused by a number of factors that can be associated with local, regional, and hemispheric climatic processes. Moreover, river flow has strong natural variability and exhibits long-term persistence which can confound the results of trend and significance tests.  相似文献   
107.
108.
The development and first tests of a novel method for the direct, sensitive, selective, and fast detection of gaseous nitric acid in air are described. The detection method is based on the laser-photolysis fragment-fluorescence (LPFF) method using 193 nm (ArF) laser light. The photolysis process yielding OH fluorescence has been thoroughly investigated. Up to now, the method has been found to be practically free of interferences. The mixing ratio for nitric acid is measured on-line with a detection limit of about 0.1 ppbv for an accumulation time of the signal of 10–15 min. A number of tests and field measurements were performed to demonstrate the feasibility and limits of the described detection method.  相似文献   
109.
A geophysical investigation of a landfill area was conducted as inextricable phase of a preliminary geotechnical, geological and hydrogeological study of the area under investigation. Electrical resistivity tomography (ERT), one of the most promising prospecting techniques mainly concerning its effective contribution to resolve several environmental problems, was applied for the geophysical modeling. ERT is a robust imaging method the theory and implementation of which are well documented in geophysical research literature. The survey site is located in Akrotirion Peninsula, about 7 km east of Chania city in Crete I. The geological setting comprises Trypalion limestones, marly limestones and a clayed weathered layer. Cavities are also identified at various scales which are recent tectono-karstic voids. Due to the above-mentioned geological singularity and in the light of the requirement for an environmentally safe construction of the landfill, an ERT survey was carried out. Specifically, seven geoelectrical tomographies were conducted reaching the prospecting depth of 36 m. The resulted images conduced, to overcome geotechnical problems since the spatial distribution of karstic voids was determined, to plan the future sites for waste disposal as the geological conditions were studied in detail and to reliably estimate the thickness of the already deposited wastes. The resulted images were confirmed using available borehole logs and ambient noise measurements.  相似文献   
110.
Anomaly analysis is used for various geophysics applications such as determination of geophysical structure's location and border detections. Besides the classical geophysical techniques, artificial intelligence based image processing algorithms have been found attractive for geophysical anomaly analysis. Recently, cellular neural networks (CNN) have been applied to geophysical data and satisfactory results are reported. CNN provides fast and parallel computational capability for geophysical image processing applications due to its filtering structure. The behavior of CNN is defined by two template matrices that are adjusted by a properly supervised learning algorithm. After training stage for geophysical data, Bouguer anomaly maps can be processed and analyzed sequentially. In this paper, CNN learning and processing capability have been improved, combining Wavelet functions and backpropagation learning algorithms. The new architecture is denoted as Wavelet-Cellular Neural networks (Wave-CNN) and it is employed to analyze Bouguer anomaly maps which are important to extract useful information in geophysics. At first, Wave-CNN performance is tested on synthetic geophysical data, which are created by a computer environment. Then, Bouguer anomaly maps of the Dumluca iron ore field have been analyzed and results are reported in comparison to real drilling results.  相似文献   
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号