Using seasonally collected data(2009-2010) from 15 sampling sites that represent first- to fifth-order streams within the Qingyi watershed,we examined the spatio-temporal patterns of fish assemblages along two longitu...Using seasonally collected data(2009-2010) from 15 sampling sites that represent first- to fifth-order streams within the Qingyi watershed,we examined the spatio-temporal patterns of fish assemblages along two longitudinal gradients to explore the effects of a large dam on fish assemblages at the watershed scale.No significant variation was observed in either species richness or assemblage structure across seasons.Species richness significantly varied according to stream order and gradient.Dam construction appeared to decrease species richness upstream substantially,while a significant decrease between gradients only occurred within fourth-order streams.Along the gradient without the large dam,fish assemblage structures presented distinct separation between two neighboring stream orders,with the exception of fourth-order versus fifth-order streams.However,the gradient disrupted by a large dam displayed the opposite pattern in the spatial variation of fish assemblages related with stream orders.Significant between-gradient differences in fish assemblage structures were only observed within fourth-order streams.Species distributions were determined by local habitat environmental factors,including elevation,substrate,water depth,current discharge,wetted width,and conductivity.Our results suggested that dam construction might alter the longitudinal pattern in fish species richness and assemblage structure in Qingyi Stream,despite the localized nature of the ecological effect of dams.展开更多
红绿灯位置是道路上行人和车辆的交会点,极大影响着道路结构和交通运行,在城市路网中起着重要的枢纽作用。针对目前红绿灯位置检测方法准确率不够高、覆盖面区域不完整等问题,提出了一种基于轨迹数据的交通灯位置检测方法。该方法基于聚...红绿灯位置是道路上行人和车辆的交会点,极大影响着道路结构和交通运行,在城市路网中起着重要的枢纽作用。针对目前红绿灯位置检测方法准确率不够高、覆盖面区域不完整等问题,提出了一种基于轨迹数据的交通灯位置检测方法。该方法基于聚类-合并-分类-合并的四级模型,首先从清理过的轨迹数据中提取隐含的车辆行驶特征,再采用具有噪声的基于密度的聚类(density-based spatial clustering of applications with noise,DBSCAN)方法得到转向和停驻两类聚类中心,对这两类聚类中心进行合并,获得红绿灯位置的候选位置;根据候选位置一定范围内的轨迹点提取该区域的车流行驶特征,然后采用梯度提升决策树(gradient boosting decision tree,GBDT)算法进行分类,最后将候选位置的正样本融合,以检测红绿灯位置。采用成都市浮动车GPS轨迹数据进行实验,检测结果的F1分数为0.947,效果优于常规的机器学习方法。实验结果表明,基于GPS轨迹数据,采用提出的四层模型能有效检测出红绿灯的位置,该模型可被用于城市大范围红绿灯位置信息的快速获取和更新。展开更多
基金Foundation items： This study was financially supported by the National Basic Research Program of China （2009CB119200） and the Natural Science Foundation of China （31071900, 31172120）
文摘Using seasonally collected data(2009-2010) from 15 sampling sites that represent first- to fifth-order streams within the Qingyi watershed,we examined the spatio-temporal patterns of fish assemblages along two longitudinal gradients to explore the effects of a large dam on fish assemblages at the watershed scale.No significant variation was observed in either species richness or assemblage structure across seasons.Species richness significantly varied according to stream order and gradient.Dam construction appeared to decrease species richness upstream substantially,while a significant decrease between gradients only occurred within fourth-order streams.Along the gradient without the large dam,fish assemblage structures presented distinct separation between two neighboring stream orders,with the exception of fourth-order versus fifth-order streams.However,the gradient disrupted by a large dam displayed the opposite pattern in the spatial variation of fish assemblages related with stream orders.Significant between-gradient differences in fish assemblage structures were only observed within fourth-order streams.Species distributions were determined by local habitat environmental factors,including elevation,substrate,water depth,current discharge,wetted width,and conductivity.Our results suggested that dam construction might alter the longitudinal pattern in fish species richness and assemblage structure in Qingyi Stream,despite the localized nature of the ecological effect of dams.
文摘红绿灯位置是道路上行人和车辆的交会点,极大影响着道路结构和交通运行,在城市路网中起着重要的枢纽作用。针对目前红绿灯位置检测方法准确率不够高、覆盖面区域不完整等问题,提出了一种基于轨迹数据的交通灯位置检测方法。该方法基于聚类-合并-分类-合并的四级模型,首先从清理过的轨迹数据中提取隐含的车辆行驶特征,再采用具有噪声的基于密度的聚类(density-based spatial clustering of applications with noise,DBSCAN)方法得到转向和停驻两类聚类中心,对这两类聚类中心进行合并,获得红绿灯位置的候选位置;根据候选位置一定范围内的轨迹点提取该区域的车流行驶特征,然后采用梯度提升决策树(gradient boosting decision tree,GBDT)算法进行分类,最后将候选位置的正样本融合,以检测红绿灯位置。采用成都市浮动车GPS轨迹数据进行实验,检测结果的F1分数为0.947,效果优于常规的机器学习方法。实验结果表明,基于GPS轨迹数据,采用提出的四层模型能有效检测出红绿灯的位置,该模型可被用于城市大范围红绿灯位置信息的快速获取和更新。
