Study of Star/Galaxy Classification Based on the XGBoost Algorithm     

《Chinese Astronomy and Astrophysics》2019,43(4):539-548

Machine learning has achieved great success in many areas today. The lifting algorithm has a strong ability to adapt to various scenarios with a high accuracy, and has played a great role in many fields. But in astronomy, the application of lifting algorithms is still rare. In response to the low classification accuracy of the dark star/galaxy source set in the Sloan Digital Sky Survey (SDSS), a new research result of machine learning, eXtreme Gradient Boosting (XGBoost), has been introduced. The complete photometric data set is obtained from the SDSS-DR7, and divided into a bright source set and a dark source set according to the star magnitude. Firstly, the ten-fold cross-validation method is used for the bright source set and the dark source set respectively, and the XGBoost algorithm is used to establish the star/galaxy classification model. Then, the grid search and other methods are used to adjust the XGBoost parameters. Finally, based on the galaxy classification accuracy and other indicators, the classification results are analyzed, by comparing with the models of function tree (FT), Adaptive boosting (Adaboost), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), Stacked Denoising AutoEncoders (SDAE), and Deep Belief Nets (DBN). The experimental results show that, the XGBoost improves the classification accuracy of galaxies in the dark source classification by nearly 10% as compared to the function tree algorithm, and improves the classification accuracy of sources with the darkest magnitudes in the dark source set by nearly 5% as compared to the function tree algorithm. Compared with other traditional machine learning algorithms and deep neural networks, the XGBoost also has different degrees of improvement.  相似文献   

基于卷积神经网络与条件随机场方法提取乡镇非正规固体废弃物     

刘懿兰  黄晓霞  李红旮  柳泽  陈崇  王新歌 《地球信息科学学报》2019,21(2):259-268

随着村镇经济建设发展,生活垃圾和工业固体废弃物造成的污染问题日益突出,已经成为制约新农村建设发展和生态文明建设的关键问题,而目前针对乡镇非正规固体废弃物的调查与统计主要依赖全国各乡镇相关部门逐级调查上报,工作量较大。本文基于高分辨率遥感影像,将深度学习模型和条件随机场模型相结合引入到乡镇固体废弃物的提取研究中,探索一种基于深度卷积神经网络的乡镇固体废弃物提取模型。由于固体废弃物在影像上表现为面积小,分布破碎等特点,为了提高工作效率,将模型特分为识别和提取2个部分：① 通过全连接卷积网络（CNN）对固体废弃物进行快速识别判断,筛选感兴趣区域影像块;② 在传统的全卷积神经网络(FCN)的基础上加入条件随机场模型(CRF)提取固体废弃物边界,提高整体分割精度。根据安徽、山西等地区相关部门上报固体废弃物堆放点以及住房与城乡建设部城乡规划管理中心进行野外检查的结果,实验最终识别精度达到86.87%以上;形状提取精度为89.84%,Kappa系数为0.7851,识别与提取精度均优于传统分类方法。同时,该方法已经逐步应用于住房和城乡建设部有关成都、兰州、河北等部分乡镇非正规固体废弃物的核查工作,取得了较为满意的结果。  相似文献   

鄂尔多斯地块周缘垂直形变场负位错反演与应变积累研究     

张希  刘立炜  郝明  白卓立 《大地测量与地球动力学》2019,39(6):551-556

利用鄂尔多斯地块及其周缘1970～2014年的垂直形变速率场资料，借助负位错反演研究该区域长期应变积累。结果表明，地块东北缘山西断陷带中北段年均能量积累增量、剪应力强度都较高，西南缘六盘山断裂与渭河断裂西段次之；山西断陷带中南段至晋陕交界处年均剪应力强度较高且显示一定程度的能量积累；西秦岭构造区尤其西秦岭北缘断裂西段、晋冀蒙交界区也反映一定程度的能量积累特性。  相似文献   

联合BDS/GPS的北斗广域差分实时电离层延迟格网改正方法研究     

房成贺  陈俊平  兰孝奇  张磊 《大地测量与地球动力学》2019,39(2):169-177

研究并实现中国区域北斗广域差分实时电离层延迟格网改正算法，针对北斗单系统实时建立电离层延迟格网的不足，提出联合BDS/GPS的北斗广域差分实时电离层延迟改正方法。比较BDS、GPS以及BDS/GPS观测的实时电离层延迟格网，并利用北斗单频单点伪距定位进行精度验证。结果表明，在有效格网点覆盖充足的地区，BDS与GPS实时电离层延迟格网定位效果相当；而联合BDS/GPS观测的实时电离层延迟格网极大提高了偏远地区的单频定位精度与可定位历元数，也使原本单系统格网覆盖充足的地区定位效果得到进一步增强和稳定。  相似文献   

碳氮稳定同位素比值在潮间带大型底栖动物组织间差异性研究   总被引：1，自引：1，他引：0  

刘博  陈琳琳  李宝泉  侯西勇  冯光海  李晓炜  李秉钧  王玉珏 《海洋学报》2019,41(4):74-82

近年来，稳定同位素技术已被广泛应用于食物网研究中。然而，生物样品取样部位对食物网构建的影响的研究相对少见。大型底栖动物作为潮间带食物网的重要组成部分，在食物网研究中有必要对其采样部位进行统一，以增加研究结果的准确性及可比较性。本研究选取黄河三角洲和烟台潮间带10种代表性的大型底栖动物，比较δ13C和δ15N在不同采样部位中的差异。结果显示，对于大多数底栖动物而言，采样部位将直接影响到δ13C和δ15N的测定结果。当底栖动物作为摄食者，需分析其食性和营养级时，应统一采用肌肉组织，如贝类的闭壳肌或足部肌肉，蟹类的鳌足肌肉，鱼类的背部肌肉；当底栖动物作为被摄食者，需分析次级消费者的食性和营养级时，除多毛类采用去除消化道内容物的体壁外，其他种类应选取整体（难以被消化的组织除外），如贝类的软体部，蟹类肌肉与鳃等的混合组织。  相似文献   

Responding to Global Warming: Mitigation Policies and Actions of Stakeholders in China’s Tourism Industry     

FENG Ling  CHEN Da  GAO Shan  WANG Yongzhi  TANG Chengcai 《资源与生态学报(英文版)》2019,10(1):94-103

China’s tourism industry has witnessed rapid progress in recent years, and is now an important part of global tourism in dealing with climate change. Within a framework of Pressure-State-Response (PSR), this paper focuses on the emission reduction pressure, carbon emission status, and responses of stakeholders in China’s tourism industry. Findings include: 1) The central government’s strategy and rapid growth of the industry scale exert rising pressure on China’s tourism to reduce carbon emissions. 2) Carbon emissions of China's tourism account for 13%-14.6% of global tourism, and about 3% of China’s emissions overall. Chinese tourists’ per capita carbon emission is lower than half of the global level. 3) The Chinese government attaches great importance to energy-saving and carbon emission reduction. In the tourism industry, documents, standards and other regulative measures have been issued to ensure that business practitioners set up green operational and managerial systems. In the field of tourism transportation, China's high-speed rail, new energy vehicles, and urban shared bicycles, have developed very rapidly in recent years, and they have effectively reduced the carbon emissions in traveling. Furthermore, this paper finds that Chinese tourists already have awareness and willingness for low-carbon tourism.  相似文献   

高分六号遥感卫星新增波段下的树种分类精度分析     

张沁雨  李哲  夏朝宗  陈健  彭道黎 《地球信息科学学报》2019,21(10):1619-1628

高分六号卫星具有覆盖广、多种分辨率、波段多的优势,能为遥感解译提供更丰富的信息。为探究高分六号卫星新增波段在森林树种识别上的应用,本文以覆盖根河市阿龙山林业局的一期高分六号宽幅影像为数据源,基于特征优化空间算法（Feature Space Optimization,FSO）和最大似然分类法,分别利用高分六号的前4个波段和所有波段（8波段）的光谱、纹理等特征进行了森林树种分类,并逐一添加新增波段特征确定了各波段的贡献率排名。结果表明：在加入了优选出的均匀性纹理、均值纹理和角二阶矩纹理3种纹理特征后,前4波段和8波段的分类精度比只基于光谱特征时的精度分别高出13.23%和24.63%;利用8波段信息比只利用前4波段在基于光谱特征上的精度高11.88%,在基于光谱+纹理特征上则高23.24%;基于8波段光谱+纹理特征的树种分类精度最高,达到68.74%,新增4波段的贡献率排名为B6>B5>B8>B7,说明新增红边波段对于本次树种分类试验的贡献率最高,能为北方树种识别提供有效帮助。  相似文献   

地质和岩土工程光电传感监测研究新进展——第六届OSMG国际论坛综述     

朱鸿鹄  施斌  张诚成 《工程地质学报》2020,28(1):178-188

第六届地质(岩土)工程光电传感监测国际论坛于2017年11月3~5日在南京大学顺利召开。本届论坛主题为"基础设施监测技术前沿及应用",共设置了35场特邀报告,来自近20个国家和地区的350余位代表参会。本届论坛显示,近几年来国内外地质和岩土工程光电传感监测领域的研究工作又取得了很多新进展:(1)光电感测解调技术不断成熟和完善,监测信噪比、空间分辨率等指标也突飞猛进;(2)国际上出现了一系列适用于地质与岩土工程监测的新型光电传感器,如聚合物光纤土工织物等;(3)光电传感器及其监测系统成为保障隧道、堤坝、核电站、桥梁等基础设施安全运营的有力工具,为这些设施的健康状态诊断和损伤识别提供了数据支撑;(4)光电传感监测技术在地质灾害监测预警中扮演了越来越重要的角色。未来的研究热点主要集中于3个方面:(1)高性价比的分布式光纤传感解调技术的研发;(2)匹配地质和岩土工程监测需求的新型光纤传感器及其布设工艺的研发;(3)基于人工智能的监测数据处理和灾害预警系统的开发。  相似文献   

自然重砂成因矿物学特征及找矿指示作用     

董国臣  李胜荣  申俊峰  董朋生  李华伟  殷国栋  汤家辉 《地学前缘》2020,27(5):171-178

自然重砂是地质体经自然风化、剥蚀、搬运、沉积等地质作用而分离出的单矿物(或矿物组合)。自然重砂矿物晶体由于仍然保留有许多矿物成因信息,包括颜色、形态、化学成分、物理性质和矿物组合等特征,因此常用于追溯源区地质体或者找矿勘查。这种方法被认为是一种经济实用的找矿方法——自然重砂测量。本文基于全国自然重砂找矿的数据资料,系统梳理了自然重砂的矿物类型、出现频率及其分布等特点,分析了自然重砂的矿物组合和成因矿物学特征,研究了自然重砂矿物的源区烙印、搬运距离及标型指示矿物组合特征,探讨了自然重砂成因矿物学研究意义及其找矿应用前景。自然重砂矿物的颜色、形态及内部结构依然保留着明确的成因矿物学信息:其颜色和晶体形态具有继承性而与其源区母体联系起来,体现源区母体的特性;其矿物组成可分出残余原生重矿物(包括造岩矿物、副矿物、矿石矿物等)和新生重砂矿物两个部分,如赤铜矿、孔雀石等反映着源区岩性体的成分或者赋存状态特征,其矿物组合也在很大程度上继承原生共生矿物而体现诸如有无矿化等意义;重砂颗粒的磨圆度、边界光滑性等表面特征反映搬运距离,有利于明确响应源区母体或者物源,而具有良好的找矿指示意义。  相似文献   

Soil resistances for laterally loaded rigid piles in multi-layered elastic soil     

Bipin K. Gupta 《Geomechanics and Geoengineering》2020,15(1):10-28

ABSTRACT

Short stubby piles like monopiles and large diameter drilled shafts undergo rigid body translation and rotation when subjected to a lateral force and/or a moment at the head. A method of analysis for these piles embedded in multi-layered elastic soil is developed using the variational principles of mechanics. Using this analysis, the soil resistance against pile movement can be rigorously related to the soil elastic constants, and the pile head displacement and rotation can be quickly calculated. The equilibrium equations for pile and soil displacements are obtained using the principle of virtual work and solved using an iterative algorithm. Pile responses obtained from the analysis match well with those obtained from three-dimensional finite element analyses in which the same inputs of loads, geometry, and material properties are given. Based on the new analysis, fitted equations for soil resistance parameters are developed, which can be used to directly calculate the pile head displacement and rotation without the use of the iterative algorithm. Numerical examples are provided that demonstrate how the method can be used to analyse practical problems.  相似文献