改进的分水岭变换算法在高分辨率遥感影像多尺度分割中的应用

由于高空间分辨率遥感影像自身的复杂性,传统的分水岭分割方法难以取得令人满意的效果。本文提出一种改进分水岭变换的高分辨率遥感影像多尺度分割方法,在抑制分水岭过分割现象的同时,还能实现对遥感影像的多尺度分割。该方法充分考虑了高分辨率遥感影像的多光谱和多尺度特性,首先,利用各向异性扩散滤波技术对影像进行平滑滤波,目的是在滤除各种噪声的同时还能保持影像的边缘特征和重要的细节信息;然后,提取影像的多尺度形态学梯度,并从梯度图像中提取标记;接着进行基于标记的分水岭变换;最后,利用改进的快速区域合并算法实现对影像的多尺度分割。实验表明,改进的算法能有效地抑制分水岭的过分割现象,对高分辨率遥感影像有较好的分割性能。     

MODIS影像的局地云量信息元数据提取算法与应用

云对于光学遥感影像质量及其反演地表参数精度有着重要影响,且其作为时空多变要素之一,在一定程度上制约了光学遥感影像的应用。对于具有2 330km的大扫描幅宽MODIS影像而言,现有的元数据标准仅能反映影像的总体云量,而无法反映云的空间分布状况,限制了MODIS数据的局地研究和应用。本文在现有遥感影像元数据标准的基础上,提出了新的元数据项--局地云量,用于反映云在条带影像中的空间分布状况,并实现在MODIS二级云掩膜条带产品(MOD35)中针对特定区域的局地云量信息提取算法。经验证,本算法能较快速和准确地提取省级行政区的局地云量信息,并可根据用户的需求进一步推广到任意指定的多边形区域,为MODIS数据在局地研究和应用提供了便利。     

高空间分辨率遥感影像多尺度分割优化组合算法

高空间分辨率遥感影像能够充分地描述地表覆盖空间异质性,可用于提取地面目标物。然而高空间分辨率在像元尺度的目标提取时易产生"椒盐效应"问题,面向对象的小尺度影像分割也受此效应影响;而大尺度的影像分割造成较小目标的遗漏。本文提出了一种针对高空间分辨率遥感影像的多尺度分割优化组合算法MOCA(Multi-scale-segmentation Optimal Composition Algorithm),基于后验概率信息熵指标选择影像中每个地面目标的最优分割尺度并集成组合,获得高空间分辨率遥感影像的多尺度分割优化组合结果。本文使用F指标和BCI(Bidirectional Consistency Index)两种指标评估地面目标物提取精度,并将MOCA与同类多尺度分割方法进行比较。实验结果表明,本文提出的MOCA算法可实现多个分割尺度的最优组合,并获得较高的地面目标物提取精度。     

遥感影像监督分割评价指标比较与分析

遥感影像空间分辨率的不断提高,一方面为使用者提供了更加丰富的地物信息,另一方面却也加大了信息准确高效提取的难度。影像分割是遥感影像目标提取的关键步骤,影像分割的效果直接影响信息提取的精度和准度。面对众多分割算法,影像分割效果评价成为遥感信息提取和目标识别研究的重点之一。面向典型目标识别问题,本文针对遥感影像监督分割评价问题,从实验的角度讨论其中具有代表性的面积匹配指数、相似尺寸指标、相关区域指标、质量合格率、欧氏距离指标1、欧氏距离指标2、面积差异指数和距离指标的实际性能与适用情况。首先,通过一系列实验测算不同分割方法下的影像与参考影像的差异情况,讨论测算结果并评估差异指标的优缺点;然后,通过对比分析与加权计算,提出了遥感影像监督分割综合评价方法,实验表明该方法在一定程度上有助于分割方法的科学选择以及影像信息提取效率的提高;最后,从评价指标与分割方法2个角度系统分析了实验结果,并指出了影像监督分割评价存在的问题以及发展趋势。     

高空间分辨率遥感影像分割定量实验评价方法综述

GEOBIA（Geographic Object-Based Image Analysis）技术针对高空间分辨率遥感影像分析的效果和精度远优于基于像元的传统方法。影像分割作为GEOBIA中的关键技术,学者们对此已经做了大量的研究,提出众多分割算法。对分割算法进行评价和分割技术本身同样重要,通过分割评价可以对分割算法的性能进行评价,比较不同分割算法的优劣,为影像选择合适的分割算法并设定合适的分割参数。影像分割的目的是为了实现影像分析操作的自动化,而主观评价法、系统评价法和分析评价法,因其无法给出客观定量指标的特点,难以应用于实时、自动化的高分辨率影像信息提取与分析系统当中。加之近年来针对分割评价方法的研究远远落后于分割算法本身,因此对定量分割评价方法进行综述对于影像分割方法及其应用研究意义重大。本文对现有的评价方法进行系统总结,建立了针对高空间分辨率遥感影像分割评价方法的分类体系。对各种方法,特别是定量的实验评价法进行对比,分析其应用范围和优劣,最后指出了高空间分辨率遥感影像分割评价未来的改进方向和应用前景。     

基于谱空间统计特征的高分辨率影像分割尺度估计

多尺度分割是面向对象遥感影像分析的关键性基础步骤,影像分割过程中尺度参数的选择直接关系到面向对象影像分析的质量和精度。本文首先从理论层面将遥感影像分割的尺度界定为基于统计的原始影像全局或局部特征的一种定量化估计,并在算法层面上将多尺度分割算法的尺度参数概括为空间尺度分割参数（类别或斑块间的空间距离）、属性尺度分割参数（类别或斑块间的属性距离）和合并阈值参数（斑块大小或斑块像元数目）;接着,提出了基于谱空间统计的高分辨率影像分割尺度估计方法;最后,以均值漂移多尺度分割算法为例,采用高空间分辨率的Ikonos、Quickbird和航空影像数据,对本文提出的基于谱空间统计的高分辨率影像分割尺度估计方法进行了验证。结果表明,该方法在一定程度上不仅避免了高分辨率遥感影像分割尺度参数选择的主观性和盲目性,还提高了面向对象影像分析的自动化程度,具有可行性和有效性。     

2015—2020年辽东湾海冰冰情时空特征及其影响因素

受冬季强寒潮侵袭,辽东湾会出现大范围结冰现象。为了分析2015—2020年辽东湾海冰冰情的变化规律与影响因素,本文选取Sentinel-1A/B数据开展辽东湾海冰监测。首先,采用巴氏距离选择最优纹理特征组合,再利用最大似然方法实现海冰分类;然后,根据上述海冰分类结果,分析海冰冰情等级、海冰外缘线、海冰面积、海冰类型和海冰结冰概率等冰情特征的变化规律;最后,研究海水深度、海温、气温和风速与海冰冰情的关系。主要结论如下：① 采用不同纹理特征组合方法和本文方法对2020年2月1日Sentinel-1B影像进行实验,结果表明本文方法的总体分类精度和Kappa系数分别为93.16%和0.85,分类精度最高。② 11月末到12月海冰类型以初生冰为主,间有灰冰;1月到2月中上旬以灰冰为主,间有初生冰和白冰;2月下旬到3月上旬的海冰类型以灰冰和初生冰为主。辽东湾内部结冰概率存在差异,北部沿岸结冰概率高于南部,东部结冰概率高于西部。辽东湾海冰冰情受海水深度、海温和气温影响明显,受风速影响较小。     

基于多任务学习的高分辨率遥感影像建筑提取

建筑物的自动提取对城市发展与规划、防灾预警等意义重大。当前的建筑物提取研究取得了很好的成果,但现有研究多把建筑提取当成语义分割问题来处理,不能区分不同的建筑个体,且在提取精度方面仍然存在提升的空间。近年来,基于多任务学习的深度学习方法已在计算机视觉领域得到广泛应用,但其在高分辨率遥感影像自动解译任务上的应用还有待进一步发展。本研究借鉴经典的实例分割算法Mask R-CNN和语义分割算法U-Net的思想,设计了一种将语义分割模块植入实例分割框架的深度神经网络结构,利用多种任务之间的信息互补性来提升模型的泛化性能。自底向上的路径增强结构缩短了低层细节信息向上传递的路径。自适应的特征池化使得实例分割网络可以充分利用多尺度信息。在多任务训练模式下完成了对遥感影像中建筑物的自动分割,并在经典的遥感影像数据集SpaceNet上对该方法进行验证。结果表明,本文提出的基于多任务学习的建筑提取方法在巴黎数据集上建筑实例分割精度达到58.8%,在喀土穆数据集上建筑实例分割精度达到60.7%,相比Mask R-CNN和U-Net提升1%~2%。     

城市建成区遥感影像边界提取与扩张分析

针对城市建成区提取过程中,仅依赖单一数据源导致精度不够的问题,本文基于面向对象分类方法和利用土地类型信息标准差统计变量,实现遥感影像中城市建城区边界的提取,并以该建成区为依据对河南省虞城县的城区空间扩张特征作了分析。实验中首先采用均值漂移分割算法对高分一号遥感影像实现分割,然后利用决策树分类算法实现土地利用类型分类,最后基于0.1 km × 0.1 km窗口统计土地利用类型标准差信息,获取建成区边界。面向实际应用,以河南省虞城县为例,采用高分一号影像获得虞城县2017年建成区数据,并基于该数据采用多个TM影像提取城区其他年份的建成区边界,实现河南省虞城县城区空间扩张特征分析。结果表明,本文方法获取的建成区边界精度较一般的监督分类提取边界有进一步的提高,精度达到89%。进而说明结合高分辨率影像提取多个年份的建成区数据的可靠性,在城市扩张研究中,对仅利用低空间分辨率提取精度不够问题和仅利用高分辨影像提取效率低等问题提供了较好的解决方案。     

综合多层优选尺度的高分辨率影像分割

采用面向对象方法处理高空间分辨率遥感影像时,影像分割质量对后续影像的信息提取结果影响很大。本文主要针对高分辨率影像分割中地物多尺度的问题,提出了一种基于多层优选尺度的高分辨率影像分割算法。该算法首先采用一系列规律变化的尺度对高分辨率影像进行多尺度分割,然后通过单分割层全局标准差的变化与尺度的关系确定一组最优分割尺度。在此基础上,通过各优选分割层之间的包含关系,局部建立多层次对象树,从整体上形成影像森林;通过局部同质性异质性综合评价指数的比较及父层光谱特征的限制来选取多层次对象树中的优势对象,从而获得最终的高分辨率影像分割结果。最后,本文分别采用了Geoeye和ZY3多光谱影像进行了2组分割实验,结果表明本文算法能有效地提高正常分割影像对象的比例。     

The progress in the study of Arctic pack ice ecology

The sea ice community plays an important role in the Arctic marine ecosystem. Because of the predicted environmental changes in the Arctic environment and specifically related to sea ice, the Arctic pack ice biota has received more attention in recent years using modem ice-breaking research vessels. Studies show that the Arctic pack ice contains a diverse biota and besides ice algae, the bacterial and protozoan biomasses can be high. Surprisingly high primary production values were observed in the pack ice of the central Arctic Ocean. Occasionally biomass maximum were discovered in the interior of the ice floes, a habitat that had been ignored in most Arctic studies. Many scientific questions, which deserve special attention, remained unsolved due to logistic limitations and the sea ice characteristics. Little is know about the pack ice community in the central Arctic Ocean. Almost no data exists from the pack ice zone for the winter season. Concerning the abundance of bacteria and protozoa, more studies are needed to understand the microbial network within the ice and its role in material and energy flows. The response of the sea ice biota to global change will impact the entire Arctic marine ecosystem and a long-term monitoring program is needed. The techniques, that are applied to study the sea ice biota and the sea ice ecology, should be improved.     

Abundance, biomass and composition of spring ice algal and phytoplankton communities of the Laptev Sea （Arctic）

Abundance,biomass and composition of the ice algal and phytoplank-ton communities were investigated in the southeastern Laptev Sea in spring 1999.Diatoms dominated the algal communities and pennate diatoms dominated the dia-tom population.12 dominant algal species occurred within sea ice and underlyingwater column,including Fragilariopsis oceanica,F.cylindrus,Nitzschiafrigida,N.promare,Achnanthes taeniata,Nitzschia neofrigida,Naviculapelagica,N.vanhoef fenii,N.septentrionalis,Melosira arctica,Clindrothecaclosterium and Pyrarnimonas sp.The algal abundance of bottom 10 cm sea icevaried between 14.6 and 1562.2×10~4 ceils l~(-1)with an average of 639.0×10~4cells l~(-1),and the algal biomass ranged from 7.89 to 2093.5μg C l~(-1)with an av-erage of 886.9μg C l~(-1),which were generally one order of magnitude higherthan those of sub-bottom ice and two orders of magnitude higher than those ofunderlying surface water.The integrated algal abundance and biomass of lower-most 20 cm ice column were averagely 7.7 and 12.2 times as those of upper 20 mwater column,respectively,suggesting that the ice algae might play an importantrole in maintaining the coastal marine ecosystem before the thawing of sea ice.Icealgae influenced the phytoplankton community of the underlying water column.However,the“seeding”of ice algae for phytoplankton bloom was negligible be-cause of the iow phytoplankton biomass within the underlying water column.     

海籍管理系统中用海变化的自动检测方法

彼此相邻的海域之间由于用海类型的不同,经常存在相互的干扰和影响。本文以国家海洋功能区划中规定的海洋功能区分类及海洋环境保护要求为基础,设计了用海变化的自动检测方法。根据不同海洋功能分区的海水水质质量的标准建立缓冲区检测算法和历史用海检测算法,缓冲区检测算法用来检测同一时期不同位置的用海是否存在冲突;历史用海检测算法用来检测同一位置不同时期的用海是否存在冲突。在两种检测算法的基础上,利用地理信息系统开发组件实现了对新加入宗海及原宗海用海变更的自动检测功能,并以上海海洋功能区划数据为测试数据,验证了作为海籍空间信息管理系统的核心功能之一,可以实现用海变化的自动检测,满足对宗海空间变化和时间变化的信息化管理。     

Polar Sea Ice Identification and Classification Based on HY-2A/SCAT Data

In this paper,a Bayesian sea ice detection algorithm is first used based on the HY-2A/SCAT data,and a backpropagation(BP)neural network is used to classify the Arctic sea ice type.During the implementation of the Bayesian sea ice detection algorithm,linear sea ice model parameters and the backscatter variance suitable for HY-2A/SCAT were proposed.The sea ice extent obtained by the Bayesian sea ice detection algorithm was projected on a 12.5 km grid sea ice map and validated by the Advanced Microwave Scanning Radiometer 2(AMSR2)15%sea ice concentration data.The sea ice extent obtained by the Bayesian sea ice detection al-gorithm was found to be in good agreement with that of the AMSR2 during the ice growth season.Meanwhile,the Bayesian sea ice detection algorithm gave a wider ice edge than the AMSR2 during the ice melting season.For the sea ice type classification,the BP neural network was used to classify the Arctic sea ice type(multi-year and first-year ice)from January to May and October to De-cember in 2014.Comparison results between the HY-2A/SCAT sea ice type and Equal-Area Scalable Earth Grid(EASE-Grid)sea ice age data showed that the HY-2A/SCAT multi-year ice extent variation had the same trend as the EASE-Grid data.Classification errors,defined as the ratio of the mismatched sea ice type points between HY-2A/SCAT and EASE-Grid to the total sea ice points,were less than 12%,and the average classification error was 8.6%for the study period,which indicated that the BP neural network classification was a feasible algorithm for HY-2A/SCAT sea ice type classification.     

Impact of the El Nino on the Variability of the Antarctic Sea Ice Extent

1　IntroductionManymeteorologistsandoceanographerspaidmuchattentiontothestudyofthemechanismofENSOformanyyears,suchasBjerknes(1 966) ,Wyrtki(1 975) ,McCreary(1 983 ) ,Philander(1 984) ,ZhangandChao(1 993 )andMcCPhaden(1 998)havemadegreatdevelopmentinthestudyofENSO .Especiallyinthe 1 990’s,withtheincreasingofthedatainthedeepocean ,thesomeonearguedthattheENSOepisodehadcloserelation shipwiththeeasterntransportationoftheanomalousseasurfacetemperatureinthewestPacific(LiandMu 1 999;Huang 2…     

Effect of compressive strength on the performance of the NEMO-LIM model in Arctic Sea ice simulation

Satellite records show that the extent and thickness of sea ice in the Arctic Ocean have significantly decreased since the early 1970s. The prediction of sea ice is highly important, but accurate simulation of sea ice variations remains highly challenging. For improving model performance, sensitivity experiments were conducted using the coupled ocean and sea ice model (NEMO-LIM), and the simulation results were compared against satellite observations. Moreover, the contribution ratios of dynamic and thermodynamic processes to sea ice variations were analyzed. The results show that the performance of the model in reconstructing the spatial distribution of Arctic sea ice is highly sensitive to ice strength decay constant (C^rhg). By reducing the C^rhg constant, the sea ice compressive strength increases, leading to improved simulated sea ice states. The contribution of thermodynamic processes to sea ice melting was reduced due to less deformation and fracture of sea ice with increased compressive strength. Meanwhile, dynamic processes constrained more sea ice to the central Arctic Ocean and contributed to the increases in ice concentration, reducing the simulation bias in the central Arctic Ocean in summer. The root mean square error (RMSE) between modeled and the CryoSat-2/SMOS satellite observed ice thickness was reduced in the compressive strength-enhanced model solution. The ice thickness, especially of multiyear thick ice, was also reduced and matched with the satellite observation better in the freezing season. These provide an essential foundation on exploring the response of the marine ecosystem and biogeochemical cycling to sea ice changes.

     

Interannual Variability and Scenarios Projection of Sea Ice in Bohai Sea Part I: Variation Characteristics and Interannual Hindcast

The Bohai Sea is one of the southernmost areas for sea ice formation in the northern hemisphere.Sea ice disasters in this body of water severely affect marine activities and the safety of coastal residents.In this study,we analyze the variation characteristics of the sea ice in the Bohai Sea and establish an annual regression model based on predictable mode analysis method.The results show the following:1)From 1970 to 2018,the average ice grade is(2.6±0.8),with a maximum of 4.5 and a minimum of 1.0.Liaodong Bay(LDB)has the heaviest ice conditions in the Bohai Sea,followed by Bohai Bay(BHB)and Laizhou Bay(LZB).Interannual variation is obvious in all three bays,but the linear decreasing trend is significant only in BHB.2)Three modes are obtained from empirical orthogonal function analysis,namely,single polarity mode with the same sign of anomaly in all of the three bays and strong interannual variability(82.0%),the north–south dipole mode with BHB and LZB showing an opposite sign of anomalies to that in LDB and strong decadal variations(14.5%),and a linear trend mode(3.5%).Critical factors are analyzed and regression equations are established for all the principal components,and then an annual hindcast model is established by synthesizing the results of the three modes.This model provides an annual spatial prediction of the sea ice in the Bohai Sea for the first time,and meets the demand of operational sea ice forecasting.     

Three Decades of Sea-Ice Variability in Jiaozhou Bay Revealed by Landsat Observations

In this study, we used Landsat images and meteorological data to examine the spatiotemporal distribution and variability of sea ice in Jiaozhou Bay(JZB) between 1986 and 2016. The results show that JZB is not always covered by sea ice in winter, but in some extreme cases, sea ice has covered more than one-third of the sea area of the bay. Sea ice in JZB has generally formed between January 1 and February 5, primarily along the coast, and gradually expanding to the central area of the bay. Both meteorological and artificial factors have played important roles in modulating the sea ice distribution. We found sea ice coverage to have been strongly correlated with the accumulated freezing-degree days nine days before the occurrence of sea ice(R2 = 0.767). North-northwest surface winds have dominated the freezing period of sea water in the JZB, and wind speed has exerted a more significant influence on the formation of sea ice when the sea ice coverage has been generally small. Additionally, artificial factors began to affect the expansion of sea ice in JZB since 2007. The construction of the Jiao-Zhou-Bay Bridge(JZBB) is believed to have retarded water flow and reduced the tidal prism, thereby leading to the formation of an ice bridge along the JZBB, which effectively prevents the southward expansion of sea ice.     

Analysis of multi-dimensional SAR for determining the thickness of thin sea ice in the Bohai Sea

Flat thin ice (<30 cm thick) is a common ice type in the Bohai Sea, China. Ice thickness detection is important to offshore exploration and marine transport in winter. Synthetic aperture radar (SAR) can be used to acquire sea ice data in all weather conditions, and it is a useful tool for monitoring sea ice conditions. In this paper, we combine a multi-layered sea ice electromagnetic (EM) scattering model with a sea ice thermodynamic model to assess the determination of the thickness of flat thin ice in the Bohai Sea using SAR at different frequencies, polarization, and incidence angles. Our modeling studies suggest that co-polarization backscattering coefficients and the co-polarized ratio can be used to retrieve the thickness of flat thin ice from C- and X-band SAR, while the co-polarized correlation coefficient can be used to retrieve flat thin ice thickness from L-, C-, and X-band SAR. Importantly, small or moderate incidence angles should be chosen to avoid the effect of speckle noise.