复杂卫星图像中的小目标船舶识别

船舶作为海上的重要目标,实现对船舶自动识别有重要的意义。针对卫星图像中云雾、海岸背景等复杂海情对船舶识别带来的干扰,以及小目标船舶高漏检率问题,本文提出一种多尺度深度学习模型训练策略,在此基础上构建了一种船舶识别的深度学习网络,该网络可分为多尺度训练、特征提取、生成目标建议区域、船舶分类这4个部分。首先,采用多尺度的训练策略,将多尺度的船舶样本送入网络中进行训练,这样在训练样本中加入了大量小目标船舶的样本,使网络充分提取到小目标船舶的特征;其次,通过卷积神经网络对目标船舶进行特征自适应提取;然后,目标区域建议网络可依据卷积神经网络提取到的特征,在图像中找到感兴趣目标区域,即框定船舶的位置;最后,通过多个全连接层的组合,将高维特征映射到一个4元组中,再运用分类函数输出每一类船舶的概率值,概率值最大的则为该船舶的类别。同时为解决云雾遮挡和海岸背景的干扰,采用了一种负样本增强学习的方法,在样本数据集中加入了大量只含有云雾和海岸背景的图片,进行负样本扩充,增强网络模型对云雾及海岸背景的特征学习能力,以此解决复杂海情的影响。实验结果表明,所提方法有效解决了复杂海情条件下的船舶识别难,以及小目标船舶识别难的问题,实现了复杂海情条件下的船舶识别。同时,与现有成熟的深度学习目标识别算法相比,本文算法的精确度和召回率分别提升了6.98%和18.17%,所训练的模型具有良好的泛化能力和鲁棒性。     

基于深度学习的目标检测定位系统设计与实现

针对传统遥感影像目标检测方法效率不高,并且无有效手段对检测信息进行管理利用的问题,提出了在B/S构架下基于深度学习的目标检测及定位方法。通过集成深度学习框架、WebGIS以及数据库,实现了集遥感影像目标检测、展示及管理于一体的目标检测定位系统,满足多用户基于前端浏览器的并发目标检测需求。利用网格划分策略,实现了基于前端的大区域范围的目标快速检测。基于某机场飞机目标及某城市区域运动场目标检测结果表明:本文设计的目标检测定位系统能够在前端实现目标快速检测定位,具有较高检测精度,并可有效管理检测信息,为深度学习循环再利用提供数据支撑。     

基于YOLOv3深度学习的海雾气象条件下海上船只实时检测

海雾气象条件下船只高精度检测识别面临较大困难,传统的目标识别、定位方法效果差强人意。作者围绕海雾气象条件下不同类型船只的实时检测问题,提出一种基于YOLOv3深度学习的实时海上船只检测新思路。首先构建清晰图片和模糊图片（海雾、雨）的判别方法,实现图片清晰度分类处理;其次为提高海雾气象条件下海上船只的实时检测精度,消除海雾遮挡对目标识别的影响,运用暗通道先验去雾方法对含有海雾的图像实行去雾;最后基于YOLOv3深度学习算法对精细处理后的图像进行船只实时检测。实验结果表明该方法能够在海雾气象条件下高效、准确地检测到船只,对海上复杂环境条件下的船只实时检测研究具有一定的理论指导意义。     

星站差分与PPP技术在深远海调查中的位置服务精度分析

利用在全球不同海域的船载GNSS进出港数据,以事后动态处理技术(PPK)的坐标计算值为评价基准,分析了Veripos星站差分系统和事后动态精密单点定位技术(PPP)在不同运动状态下的位置服务精度。结果表明,Veripos与PPP在低动态下的平面与高程位置服务精度可优于6.2 cm和14.9 cm,在常规动态下平均精度优于20 cm;PPP与Veripos坐标分量平均互差小于15 cm,标准差小于23 cm,在深远海调查中PPP技术可作为星站差分系统的有效补充和特殊情况下的位置服务替代。     

基于机器学习的海洋环境预报订正方法研究

结合中尺度数值模式 WRF 预报数据和 ERA5 再分析资料,利用机器学习方法对 WRF 预报场的风场、温度、气压进行预报订正。采用 ERA5 作为真值,与原始 WRF 预报相比,利用随机森林模型可以将预报结果整体均方根误差降低 44%以 上,利用深度神经网络模型可以将预报结果整体均方根误差降低 34%以上。通过随机森林模型实验得到不同输入特征对预报要素的影响程度,分析了关键的预报订正因子。     

基于双隐层ANN模型的叶绿素a浓度智能预报方法

利用2019年5月WZ02生态浮标监测数据,建立了两种不同隐层人工神经网络(ANN)模型的叶绿素a(Chl-a)智能预报方法,并对单隐层和双隐层模型的预测结果做了对比.结果表明:双隐层结构预测结果精度更高,泛化能力更强,一定程度上说明了深层学习比浅层学习对信息的主要特征提取能力更有优势.同时,对数据样本集合进行了系统预...     

贵州息烽温泉控热推覆构造的发现及成因研究

根据近年来磷矿深部勘探资料,对贵州息烽温泉的构造地质条件及地热特征开展研究,获得了有关温泉地质成因的新认识。(1)由洋水背斜、安清断层(F₁)和温泉断层(F₂)组成的叠瓦状逆冲推覆构造体系,构成了息烽温泉地热水系统的主体及边界,该控热构造体系内地表构造与地腹构造并不统一,老地层盖在新地层上,构成顶、底两层叠置的多个热储单元,这些热储单元是地热水形成和运移的良好场所。(2)安清断层(F₁)使多个热储单元直接接触,从而具有密切的水力联系,为导水、导热的良好通道;温泉断层(F₂)构成了息烽温泉区地热水系统的边界。(3)大气降水沿断裂通道下渗至深部的碳酸盐岩热储层中,吸收热量形成地热水,在高温高压条件下,向洋水背斜核部进行迁移,受上覆寒武系碎屑岩盖层的保护和温泉断层的限制,形成洋水背斜西翼和北端的地热资源有利富集区,最后受分支断层阻挡而自然岀露。     

Strain analysis of a seismically imaged mass‐transport complex,offshore Uruguay

Strain style, magnitude and distribution within mass‐transport complexes (MTCs) are important for understanding the process evolution of submarine mass flows and for estimating their runout distances. Structural restoration and quantification of strain in gravitationally driven passive margins have been shown to approximately balance between updip extensional and downdip contractional domains; such an exercise has not yet been attempted for MTCs. We here interpret and structurally restore a shallowly buried (c. 1,500 mbsf) and well‐imaged MTC, offshore Uruguay using a high‐resolution (12.5 m vertical and 15 × 12.5 m horizontal resolution) three‐dimensional seismic‐reflection survey. This allows us to characterise and quantify vertical and lateral strain distribution within the deposit. Detailed seismic mapping and attribute analysis shows that the MTC is characterised by a complicated array of kinematic indicators, which vary spatially in style and concentration. Seismic‐attribute extractions reveal several previously undocumented fabrics preserved in the MTC, including internal shearing in the form of sub‐orthogonal shear zones, and fold‐thrust systems within the basal shear zone beneath rafted‐blocks. These features suggest multiple transport directions and phases of flow during emplacement. The MTC is characterised by a broadly tripartite strain distribution, with extensional (e.g. normal faults), translational and contractional (e.g. folds and thrusts) domains, along with a radial frontally emergent zone. We also show how strain is preferentially concentrated around intra‐MTC rafted‐blocks due to their kinematic interactions with the underlying basal shear zone. Overall, and even when volume loss within the frontally emergent zone is included, a strain difference between extension (1.6–1.9 km) and contraction (6.7–7.3 km) is calculated. We attribute this to a combination of distributed, sub‐seismic, ‘cryptic’ strain, likely related to de‐watering, grain‐scale deformation and related changes in bulk sediment volume. This work has implications for assessing MTCs strain distribution and provides a practical approach for evaluating structural interpretations within such deposits.     

深水等深流沉积中记录的米兰科维奇特性——以陕西富平地区上奥陶统赵老峪组为例

陕西富平地区上奥陶统赵老峪组为深水盆地边缘至深水斜坡沉积环境,发育深水原地沉积、等深流沉积及重力流沉积。以野外实测剖面、岩相特征等资料为基础,结合米兰科维奇理论、天文地层学、古海洋学以及古气候学,系统研究该地区等深流强度与古气候、海平面、古盐度的关系。结果表明:Fischer图解可以客观、准确地反映深水沉积地层相对海平面变化;深水沉积及等深流沉积中不同级次旋回垂向组合关系记录了米兰科维奇特性,即小级别旋回在垂向上有序地按照1∶4或者1∶5的形式叠加构成较大级别旋回;等深流强度受米兰科维奇旋回影响,呈多周期弱→强→弱递变;等深流强化周期约为0.1Ma。     

Bathymetry,substrate and fishing areas of Southeast Atlantic high-seas seamounts

Most of the Southeast Atlantic Ocean is abyssal, and global bathymetries suggest that only ～3.2% of the areas beyond national jurisdiction (ABNJ; also known as the high seas, as defined in the United Nations Convention on the Law of the Sea [UNCLOS]) are shallower than 2 500 m. This study mapped bathymetry and characterised substrates in selected seamount summit areas, including several that have been or may become fishing areas. The southernmost location, the Schmitt-Ott Seamount, has exposed volcanic bedrock with surrounding flats covered by thin biogenic sediments and/or coral rubble that appears ancient. At Wüst, Vema, Valdivia and Ewing seamounts the basaltic base appears to be overlain by coral caps and other coral substrates (sheets, rubble). Adjacent summit plains have biogenic sediments of varying thickness. Vema has a flat, roughly circular summit, <100 m deep, with the shallowest point being a 22-m-deep summit knoll; the upper slopes have ancient coral framework, but the summit has a mixture of coralline and volcanic rock and coarse sediments, including extensive areas with coralline algae and kelp forests. Valdivia Bank is a 230-m-deep, flat, rocky area (～11 × 5 km), protruding steeply from the extensive multi-summit Valdivia subarea of the Walvis Ridge. The distribution of past fisheries in the Convention Area of the South East Atlantic Fisheries Organisation (SEAFO) was considered in relation to the new information on bathymetry and substrate.