POS数据辅助的航空影像变化检测方法研究

提出了一种利用机裁定位定向系统（POS）数据辅助航空影像进行影像匹配和变化检测的方法。首先利用带POS数据的老影像解求新影像的外方位元素，然后在老立体影像上提取特征点，根据前方交会和共线条件方程得到新影像上同名点的近似位置，再与新影像进行匹配，寻找匹配不好的点作为变化区域的初始位置。以此为基础选择精检测窗口，进行边缘提取和跟踪，并进行链码匹配，最终确定发生变化的区域。试验证实，本文方法是可行的。     

Conservation of wetlands especially as waterfowl habitat in northeast China

Ｗｅｔｌａｎｄ,ｗｉｔｈｖｅｒｙｈｉｇｈｂｉｏｌｏｇｉｃａｌｐｒｏｄｕｃｔｉｖｉｔｙ,ｔｒｅｍｅｎｄｏｕｓｒｅｓｏｕｒｃｅｐｏｔｅｎｔｉａｌａｎｄｅｎｖｉｒｏｎｍｅｎｔｆｕｎｃｔｉｏｎ,ｉｓａｓｐｅｃｉａｌｅｃｏｓｙｓｔｅｍｉｎｔｈｅｅａｒｔｈａｎ...     

A Linked Physical and Biological Framework to Assess Biogeochemical Dynamics in a Shallow Estuarine Ecosystem

The littoral zone of Chesapeake Bay contains a mosaic of shallow vegetated and nonvegetated habitats with biotic components that are sensitive to changes in biological and physical driving factors. Static and dynamic modelling frameworks provide an integrative way to study complex hydrodynamic and biogeochemical processes in linked estuarine habitats. In this study we describe a spatial simulation model developed and calibrated relative to a specific littoral zone, estuarine ecosystem. The model consisted of four distinct habitats that contained phytoplankton, sediment microalgae, Zostera marina (eelgrass), and Spartina alterniflora. There was tidal exchange of phytoplankton, particulate and dissolved organic carbon and dissolved inorganic nitrogen between the littoral zone ecosystem and the offshore channel. Physical exchange and biogeochemical transformations within the habitats determined water column concentrations in each habitat. Predicted subtidal water column concentrations and Z. marina and S. alterniflora biomass were within the variability of validation data and the predicted annual rates of net primary production were similar to measured rates. Phytoplankton accounted for 17%, sediment microalgae 46%, the Z. marina community 24% and S. alterniflora 13% of the annual littoral zone primary production. The linked habitat model provided insights into producer, habitat and ecosystem carbon and nitrogen properties that might not have been evident with stand-alone models. Although it was an intra-ecosystem sink for particulate carbon, the seagrass habitat was a DOC source and responsible for over 30% of the littoral zone carbon and nitrogen primary production. The model predicted that the Goodwin Islands littoral zone was a sink of channel derived POC, but a source of DOC to the surrounding estuary. The framework created in this study of estuarine ecosystem dynamics is applicable to many different aquatic systems over a range of spatial and temporal scales.     

A review of factors influencing the availability of dissolved oxygen to incubating salmonid embryos

Previous investigations into factors influencing incubation success of salmonid progeny have largely been limited to the development of empirical relationships between characteristics of the incubation environment and survival to emergence. It is suggested that adopting a process‐based approach to assessing incubation success aids identification of the precise causes of embryonic mortalities, and provides a robust framework for developing and implementing managerial responses. Identifying oxygen availability within the incubation environment as a limiting factor, a comprehensive review of trends in embryonic respiration, and processes influencing the flux of oxygenated water through gravel riverbeds is provided. The availability of oxygen to incubating salmonid embryos is dependent on the exchange of oxygenated water with the riverbed, and the ability of the riverbed gravel medium to transport this water at a rate and concentration appropriate to support embryonic respiratory requirements. Embryonic respiratory trends indicate that oxygen consumption varies with stage of development, ambient water temperature and oxygen availability. The flux of oxygenated water through the incubation environment is controlled by a complex interaction of intragravel and extragravel processes and factors. The processes driving the exchange of channel water with gravel riverbeds include bed topography, bed permeability, and surface roughness effects. The flux of oxygenated water through riverbed gravels is controlled by gravel permeability, coupling of surface–subsurface flow and oxygen demands imposed by materials infiltrating riverbed gravels. Temporally and spatially variable inputs of groundwater can also influence the oxygen concentration of interstitial water. Copyright © 2006 John Wiley & Sons, Ltd.     

�����쳣ƥ�丨����������ģ�͵��Ż�

???????????????1??????????????????????????????????????????????????λ?ù???????????????????????λ??????????????е????????????????е????????????????????????????????????????????????????????????????????????????????????仯?????????и???????Ч????     

Shell‐shape and morphometric variability in Mytilus galloprovincialis from micro‐tidal environments: responses to different hydrodynamic drivers

Tidal conditions differently influence inter‐tidal organisms in terms of general physiological and metabolic responses. In this study we investigated the morphological response in shells of Mytilus galloprovincialis native to different micro‐tidal coastal environments in the Northern Adriatic Sea. Our purpose was to highlight the ecophenotypic variability across tidal levels and to elucidate how tidal currents and waves produced by anthropogenic activities may play a part in modulating shell morphology. Three sampling sites were selected: an open‐sea area 15 km off‐shore and two sites within the lagoon of Venice, the first near one of its three inlets, and the other one in the proximity of Venice city centre. At each sampling site, organisms were seasonally collected at different depths within their vertical zonation, either in the inter‐tidal zone – i.e. at both the highest and lowest tide zonation limits, and subtidally. The mussel shells were analysed by investigation of their morphometric relationships (height/length and width/length ratios) and by elliptic Fourier analysis of the shell contours. Shell thickness and condition index were also evaluated for a better comprehension of energy allocation/partitioning. Estimates based on long‐term measurements, visual observation, wind statistics and wave growth laws allowed an evaluation of the forces acting on shells. At the open‐sea site, the observed phenotypic variability of both shell shape and thickness was clearly related to the tidal vertical zonation. At the two lagoon sites, the currents generated by tidal flow through the inlet and the waves caused by the frequent passage of boats influenced both shell shape and thickness. A trade‐off between protection and growth was apparent along the tide gradient, as emphasized by the differences in the partitioning and allocation of energy between shell and flesh production.     

海表水温变动对东、黄海鲐鱼栖息地分布的影响

海表水温(SST)通常是表征鱼类栖息地分布的主要指标。本文根据1999—2007年我国大型灯光围网的鲐鱼生产统计数据,结合海洋遥感获得的SST,分析了渔汛期间鲐鱼栖息地的适宜SST范围,探讨了SST变动情况下鲐鱼栖息地的变化趋势。研究结果表明,东、黄海鲐鱼7—12月的适宜SST范围为15~30℃。根据政府气候变化专门委员会(IPCC)第四份评估报告,本文拟定4种SST上升的情况,即(1)每月平均SST+0.5℃;(2)每月平均SST+1℃;(3)每月平均SST+2℃;(4)每月平均SST+4℃。结果显示,东、黄海鲐鱼的潜在栖息有明显向北移动的趋势,并且栖息地面积逐渐减小。研究认为,全球气候变化引起的SST上升,可能会对近海鲐鱼栖息地造成严重的影响。     

基于不同权重的栖息地指数模型预报阿根廷滑柔鱼中心渔场

本文根据2003-2009年1-5月和2011年1-5月西南大西洋海域阿根廷滑柔鱼(Illex argentinus)的生产数据,结合遥感获得的海表面温度(SST)和海表面高度(SSH)数据,利用不同权重的栖息地指数模型来预报阿根廷滑柔鱼的中心渔场。采用外包络法,利用作业次数与SST、SSH建立适应性指数(SI)模型,依据作业次数比重和产量比重来比较不同权重的算术加权模型(AWM),从而筛选出最佳模型,并对最佳模型进行验证。结果显示,确定AWM(a=0.3,SST权重为0.3,SSH的权重为0.7)为最佳模型,当栖息地适应性指数(HSI)大于0.6时,作业次数的比重为93.23%,产量比重为89.28%,当HSI小于0.4时,作业次数的比重为2.12%,产量比重为3.35%。利用2011年1-5月的生产数据和环境数据对AWM(a=0.3)进行验证,结果显示,在HSI大于0.6的海域,各月作业次数比重均在91%以上,产量比重均在95%以上。研究表明,在阿根廷滑柔鱼渔场形成中SSH比SST更为重要,基于SST和SSH的AWM(a=0.3)能够较好地预测西南大西洋阿根廷滑柔鱼的中心渔场。     

顾及几何特征和拓扑连续性的道路网匹配方法

已有的道路网匹配方法需要一定的人为干预和控制,仍无法实现匹配自动化,而且对于具有一定尺度差异的数据而言,受不相关道路对象的影响,匹配结果中会存在较多的误匹配。为此,本文提出一种顾及几何特征和拓扑连续性的由粗匹配到精匹配的分层匹配策略。粗匹配阶段,利用短边中位数Hausdorff距离计算匹配对象的相似度,再以道路网弧段端点的近邻分析结果作为匹配阈值,判断匹配对象是否为同名实体;精匹配阶段,将粗匹配阶段评价结果中离散的同名实体构建为完整的道路Stroke,即以追踪的方式剔除误匹配和添加漏匹配。利用宿城区不同尺度的道路网对所提算法进行验证,结果表明,该算法具有较好的匹配效果,能够实现匹配自动化。     

多匹配策略融合的无人机影像匹配方法

针对AKAZE算法在无人机影像匹配过程中存在的匹配精度低和稳定性较差问题,本文提出一种基于多匹配策略融合的改进影像匹配方法。该方法首先对影像降采样并利用AKAZE算法检测多尺度特征。然后采用一种稳定的RootSIFT描述符进行特征描述。其次,融合最近邻距离比值、双向匹配和余弦相似度约束匹配策略进行特征匹配以降低误匹配率。最后,采用随机抽样一致性(RANSAC)算法确定最终的特征对应关系,并求得几何变换模型。实验结果表明,该方法在获得更多正确匹配点对的同时具有较高的匹配正确率和精度,能够更好适用于无人机影像匹配。