GF-1 WFV图像经验模分解的光谱保真性与水深遥感探测

水深是海洋环境的重要参数之一,水深遥感反演是水深测量的一种重要手段。经验模分解（EMD）具有剔除小尺度波浪信息,留下大尺度水下地形信息的特性。本文利用EMD对高分一号卫星宽幅影像进行尺度变换,使用光谱相关系数、光谱角、光谱偏差和光谱相对偏差等评价指标,对剩余层图像进行光谱保真性分析;利用改进的对数转换比值模型对原始影像和剩余层图像进行水深反演,并进行相关性分析与精度评价。研究结果表明:（1）评价指标显示EMD变换后影像具有相当的保真性;空间断面分析表明EMD去除了小尺度的噪声信息,保留了水下地形变化信息。（2）经均匀分布的检查点验证,两区域的原图像反演水深和实测水深的相关性较好,相关系数达0.75以上,且两种波段组合的MAE和MRE均不超过2.42 m和8.5%。（3）对EMD的全部10层进行水深反演,蓝绿波段的MAE和MRE均不高于1.62 m和5.8%;绿红波段的MAE和MRE均不高于1.93 m和6.9%。（4）对于不同的波段组合,蓝绿波段组合在各剩余层的水深反演效果明显优于绿红波段,经EMD后的水深反演效果明显提高。（5）20~30 m水深段的反演精度整体要高于30~40 m,该模型应用于较浅水深段更具优势。     

高分六号遥感影像水深反演能力评价

水深是重要的海洋要素,水深遥感反演是获取浅水水深的重要手段。当前水深遥感反 演应用以国外卫星数据为主,国产卫星数据的研究和应用较少。本文针对国产高分六号卫星 （GF-6） 数据,以三亚南山港为研究区域,分别建立单波段回归模型、双波段比值模型、多波段 回归模型,进行多光谱影像的水深反演能力研究,并与国外主流哨兵2 号卫星（Sentinel-2） 数 据进行实验比较。实验结果表明：GF-6 遥感影像具有较好的浅水水深反演能力和一定的反演精度,各波段水深探测能力依次为：绿波段跃蓝波段跃红波段跃近红外波段,反演方法效果依次为：多波段模型跃双波段模型跃单波段模型。相较于Sentinel-2 数据,GF-6 数据水深反演精度与其一致,这表明GF-6 影像具备替代国外遥感数据进行水深反演的能力和大规模应用的潜力。本文针对GF-6 影像水深反演能力的研究方法和分析,结果将为国产高分系列卫星数据的水深反演研究和应用提供有益的参考。     

基于半经验模型的水深反演及不同水深范围的误差分析

基于半经验遥感模型,开展广东省雷州湾SPOT-5影像不同水深范围的遥感反演及误差分析,给出不同模型对应的最佳水深范围和不同水深范围的最佳反演模型,结果如下：红光单波段模型、红光-绿光双波段模型、绿光-红光-近红外三波段模型均在2～5m水深范围内反演误差最小,分别为16.7%,13.2%和17.1%。0～2m水深范围内,红光-近红外双波段模型反演误差最小,为34.3%;2～5m水深范围内,红光-绿光双波段模型反演误差最小,为13.2%;5～10m水深范围内,红光-近红外双波段模型反演误差最小,为19.3%;10～20m水深范围内,三波段模型反演误差最小,为31.5%。     

基于WorldView-2遥感影像反演浅海水深过程中太阳耀斑的去除方法

基于Hedley等的方法对WorldView-2遥感影像反演岛礁水深时的太阳耀斑进行去除,在技术实现过程中根据WorldView-2波段设置的特殊性做一定的改进,将8波段分为两组,其中蓝、绿、红波段依据近红外1波段来去耀斑,海岸、黄、红边波段依据近红外2波段来去耀斑,在具体应用中完整实现该理论方法。结果表明,去耀斑后无论是对水深反演结果的精度上还是从影像直观视觉上都有提高,该方法有利于WorldView-2影像处理中耀斑去除的常规化应用。     

一种改进的地理加权回归模型水深反演方法

利用多光谱卫星遥感影像反演浅海水深是水深测量的一种重要方式.提出一种基于主成分分析的地理加权回归模型(PCA-GWR),采用WorldView-2多光谱卫星遥感影像数据,对经过数学变换后的波段反射率数据先进行主成分分析,将得到第一主成分量进行地理加权回归分析,并与双波段比值模型、多波段线性模型和地理加权回归模型(GWR...     

基于梯度提升决策树算法的水深反演研究

传统的水深测量方法多通过舰载声纳实地探测的方法,灵活性较差且水深资料更新周期长,并且在某些海域,船只往往难以靠近从而无法完成测量。本文使用七连屿海域附近的WorldView-2多光谱遥感影像构建了基于梯度提升决策树(Gradient Boosting Decision Tree, GBDT)算法的水深反演模型,并利用单波束与人工测量相结合的水深数据,与传统的单波段模型、双波段模型以及BP神经网络水深反演模型的水深数据进行了水深反演精度对比。结果表明,在0～20 m深海域,GBDT模型反演精度高于其他模型,且更符合实际水深,其检验点的R²为0.9664, RMSE为0.94 m, MAE为0.75 m, RME为19%。     

基于神经网络技术的遥感水深反演模型研究

利用Landsat7 ETM+遥感影像反射率和实测水深值之间的相关性,选取了相关性较好的ETM1、ETM2、ETM3、ETM4、ETM3/ETM2等5个水深反演因子,建立了BP神经网络水深反演模型。为充分体现BP神经网络模型的优越性,利用SPSS软件建立了单波段、波段比值、多波段三种不同的线性回归模型。通过对比发现,具有很好的自适应能力和非线性映射能力的BP神经网络模型在处理遥感水深反演问题上比传统的线性模型效果更好。     

基于耀斑改正的多光谱水深反演方法研究

利用WorldView-2四波段卫星数据和电子海图数据,基于改进的耀斑改正算法和双波段比值算法,反演获得了3处典型海域的浅水水深。通过不同海域、不同耀斑条件下水深反演实验,探讨了RED,NIR波段在典型四波段水深反演中的作用和影响,发现在双波段比值法水深反演中,引入RED+NIR波段进行耀斑改正处理,可以增加珊瑚、海藻等绿色物质覆盖海底的反射率,有效地提高该类海域的水深反演精度。基于耀斑改正的多光谱水深反演方法,适用于中轻度耀斑条件下,水质较清澈的浅海水深反演,可在国内外典型四波段卫星数据水深反演中推广应用。     

中国北方沿海典型澙湖—山东月湖水深遥感反演

澙湖是中国北方重要的滨海生态系统之一,水深较浅、透明度较低、底质类型复杂、人类干扰频繁等特点导致澙湖水深资料长期缺乏或存在较大误差,极大地制约了澙湖生态系统的研究和保护。本研究在现场实测水深的基础上,使用IKONOS高分辨率卫星影像,将山东荣成月湖水域划分为浅水区、植被区和深水区三类分别进行水深反演。研究表明红光波段(Band3)对水深的敏感性最高,分区建立线性回归模型反演水深的精度高于不分区的精度,多波段组合模型的反演精度最高,植被对水深反演的精度影响较大。反演结果表明,月湖水深最深处达271.23cm,水深反演的平均相对误差为13.16%。     

控制点与检查点数量和比例对水深遥感反演精度的影响分析

在水深遥感反演半经验模型中,水深控制点和检查点选取是一项非常重要的工作,前者用于建立实测水深值与遥感影像灰度值之间的数量关系,后者用于评价水深反演精度。但前人在相关研究中并未就控制点和检查点的选取数量和比例给出有实验依据的建议,亦未开展其对水深反演精度的影响分析。针对这个问题,作者使用6020个实测水深点和World View-2、GF-1 WFV、Landsat8 OLI 3种多光谱影像,基于三波段水深反演模型开展了81组实验,分析比较了不同数量和比例的控制点与检查点对反演结果的影响,评价过程利用平均绝对误差(Mean Absolute Error,MAE)、平均相对误差(Mean Relative Error,MRE)和反映实测水深与反演水深相关性的决定系数R2等3种参数。结果表明:(1)当控制点数量达到31个时,水深反演精度即趋于稳定;(2)检查点数量在30个时,其评价指标已可以代表模型反演精度;(3)控制点和检查点的数量比例对反演精度并无影响,但控制点选取过多则会产生反演精度降低的现象,采用人工选取检查点时剔除浅水区的异常点即可有效避免这一问题。     

A detection model of underwater topography with a series of SAR images acquired at different time

underwater topography is one of oceanic features detected by Synthetic Aperture Radar. Underwater topography SAR imaging mechanism shows that tidal current is the important factor for underwater topography SAR imaging. Thus under the same wind field condition, SAR images for the same area acquired at different time include different information of the underwater topography. To utilize synchronously SAR images acquired at different time for the underwater topography SAR detection and improve the precision of detection, based on the detection model of underwater topography with single SAR image and the periodicity of tidal current, a detection model of underwater topography with a series of SAR images acquired at different time is developed by combing with tide and tidal current numerical simulation. To testify the feasibility of the presented model, Taiwan Shoal located at the south outlet of Taiwan Strait is selected as study area and three SAR images are used in the underwater topography detection. The detection results are compared with the field observation data of water depth carried out by R/V Dongfanghong 2, and the errors of the detection are compared with those of the single SAR image. All comparisons show that the detection model presented in the paper improves the precision of underwater topography SAR detection, and the presented model is feasible.     

Depth dependence of noise resulting from ship traffic and wind

Under conditions of distantly generated noise, the noise level is found to decrease with depth in the mid-northeastern Pacific. These data show a decrease in noise level greater than 25 dB between critical depth and the ocean bottom. A result of this decrease is that locally wind-generated noise can be detected on near-bottom receivers for wind speeds less than 10 kn. It is shown that the noise level generated form local sources such as wind and nearby shipping is almost independent of receiver depth. The differences in spectra shape between the distant shipping noise and wind-generated noise and the low noise levels detected near the ocean bottom allow the measurement in the frequency band at 200-500 Hz of local wind noise level for wind speeds less than 10 kn     

Crucial physical characteristics of sea ice in the Arctic section of 143°-180°W during August and early September 2008

潮流是水下地形SAR成像的决定性因素之一,潮流的周期性变化使水下地形SAR成像与潮流场有密切的关系,同一水下地形不同时刻SAR影像不同。以台湾浅滩为例,基于5景不同时相的ERS-2 SAR影像和台湾浅滩实测水深数据,开展了水下地形SAR成像与潮流场的相关性分析。为了排除SAR影像自身成像质量对研究的影响,首先对5景SAR影像的成像质量进行评价,然后通过潮汐潮流数值计算确定SAR成像时刻的潮流场以及基于QuikScat散射计数据确定SAR成像时刻的海面风向。在上述基础上开展了SAR影像后向散射系数与实测水深的相关性分析,总结出水下地形SAR成像与潮流场的关系,结果表明,潮流场与水下地形SAR成像密切相关,当潮流流向与水下地形梯度方向一致或接近时,水下地形SAR成像效果最佳;对于同一水下地形,流向相反的两种潮流对应的SAR影像后向散射系数恰好相反,即SAR影像中的明暗条纹不同。相对于潮流流向的顺流与逆流风向的不同对水下地形SAR成像无影响。     

水下地形SAR成像与潮流场的相关性分析示例

潮流是水下地形SAR成像的决定性因素之一,潮流的周期性变化使水下地形SAR成像与潮流场有密切的关系,同一水下地形不同时刻SAR影像不同。以台湾浅滩为例,基于5景不同时相的ERS-2 SAR影像和台湾浅滩实测水深数据,开展了水下地形SAR成像与潮流场的相关性分析。为了排除SAR影像自身成像质量对研究的影响,首先对5景SAR影像的成像质量进行评价,然后通过潮汐潮流数值计算确定SAR成像时刻的潮流场以及基于QuikScat散射计数据确定SAR成像时刻的海面风向。在上述基础上开展了SAR影像后向散射系数与实测水深的相关性分析,总结出水下地形SAR成像与潮流场的关系,结果表明,潮流场与水下地形SAR成像密切相关,当潮流流向与水下地形梯度方向一致或接近时,水下地形SAR成像效果最佳;对于同一水下地形,流向相反的两种潮流对应的SAR影像后向散射系数恰好相反,即SAR影像中的明暗条纹不同。相对于潮流流向的顺流与逆流风向的不同对水下地形SAR成像无影响。     

Multi-scale variability of beach profiles at Duck: A wavelet analysis

Beach profiles have been observed to change over a range of spatial and temporal scales; however techniques for quantifying this variability have not been fully established. In this paper, a wavelet technique is introduced as a method to study the multi-scale variability of beach profiles. The beach profile data comprising a 22-year time series surveyed at the US Army Corps of Civil Engineers Field Research Facility (FRF) at Duck are analysed using the adapted maximal overlap discrete wavelet transform (AMODWT). The analysis successfully identifies strong local features in the variability of beach profiles in time and space separately that cannot be isolated by traditional statistical methods. The analysis of spatial wavelet variances provides a new means of investigating the depth of closure. Analysis of variances by temporal scales shows that the combined effects of several temporal scales with one or two dominant scales can be seen at particular points across profiles whilst the dominant temporal scales are different at different portions of the profiles. The method allows for the extremely nonstationary behaviour of beach profile to be analysed into separate frequency bands that can facilitate the interpretation of morphological changes in terms of physical processes.     

海洋环境噪声垂直分布测试和分析

采用船舷法对某海域海洋环境噪声垂直分布进行了测量.数据处理与分析结果表明,在6种接收深度下,当地的海面风生破碎波浪噪声对环境噪声有显著贡献.给出了所测海域环境噪声在0.1～20.0kHz频段的宽带声级和接收深度以及多种频率谱级与风速的对数之间的关系.1.0～4.0kHz频段的谱级与风速的对数呈良好的线性关系,且基本上不受接收深度的影响.     

南海岛礁遥感影像大气校正算法研究

针对目前传统大气辐射传输模型使用时难以确定气溶胶光学厚度的问题,提出一种仅依靠影像自身信息的大气校正算法,该算法针对南海水域清澈的特点,将暗像元法和大气辐射传输模型相结合,通过暗像元匹配和迭代运算确定气溶胶.实验结果表明:该算法与Sen2cor算法相比大气校正后的地物剖面光谱曲线基本一致,校正后影像能够反映出水深与蓝绿...     

基于潜标测量的海洋环境噪声谱特性分析

利用海洋环境噪声测量潜标系统对南海典型海域开展了为期3个月的海洋环境噪声测量,16通道海洋环境噪声测量系统每小时测量两分钟噪声信号。数据处理结果表明,800~5 000Hz范围内,噪声谱与风速相关性最好,且风速越大相关性越好,噪声谱与风速的相关性好于与浪高的相关性。风关噪声谱级在海水中部基本不随接收深度发生变化,但由于测量水听器阵长度未能覆盖整个水深,因此未给出海面和海底处谱级变化规律。在400Hz以上的高频段整个风速范围内噪声谱级都随风速发生变化,且噪声谱级与对数风速具有很好的线性关系。     

Large aperture seismic imaging at a convergent margin: Techniques and results from the Costa Rica seismogenic zone

In March and April 1995 a cooperative German, Costa Rican, and United States research team recorded onshore-offshore seismic data sets along the Pacific margin of Costa Rica using the R/V Ewing. Off the Nicoya Peninsula we used a linear array of ocean bottom seismometers and hydrophones (OBS/H) with onshore seismometers extending across much of the isthmus. In the central area we deployed an OBS/H areal array consisting of 30 instruments over a 9 km by 35-km area and had land stations on the Nicoya Peninsula adjacent to this marine array and also extending northeast on the main Costa Rican landmass. Our goal in these experiments was to determine the crustal velocity structure along different portions of this convergent margin and to use the dense instrument deployments to create migrated reflection images of the plate boundary zone and the subducting Cocos Plate. Our specific goal in the central area was to determine whether a subducted seamount is present at the location of the 1990, M 7 earthquake off the Nicoya Peninsula and can thus be linked to its nucleation. Subsequently we have processed the data to improve reflection signals, used the data to calculate crustal velocity models, and developed several wide-aperture migration techniques, based on a Kirchhoff algorithm, to produce reflection images. Along the northern transect we used the ocean bottom data to construct a detailed crustal velocity model, but reflections from the plate boundary and top and bottom of the subducting Cocos plate are difficult to identify and have so far produced poor images. In contrast, the land stations along this same transect recorded clear reflections from the top of the subducting plate or plate boundary, within the seismogenic zone, and we have constructed a clear image from this reflector beneath the Nicoya shelf. Data from the 3-D seismic experiment suffer from high-amplitude, coherent noise (arrivals other than reflections), and we have tried many techniques to enhance the signal to noise ratio of reflected arrivals. Due to the noise, an apparent lack of strong reflections from the plate boundary zone, and probable structural complexity, the resulting 3-D images only poorly resolve the top of the subducting Cocos Plate. The images are not able to provide compelling evidence of whether there is a subducting seamount at the 1990 earthquake hypocenter. Our results do show that OBS surveys are capable of creating images of the plate boundary zone and the subducting plate well into the seismogenic zone if coherent reflections are recorded at 1.8 km instrument spacing (2-D) and 5 km inline by 1 km crossline spacing for 3-D acquisition. However, due to typical high amplitude coherent noise, imaging results may be poorer than expected, especially in unfavorable geologic settings such as our 3-D survey area. More effective noise reduction in acquisition, possibly with the use of vertical hydrophone arrays, and in processing, with advanced multiple removal and possibly depth filtering, is required to achieve the desired detailed images of the seismogenic plate boundary zone.