Study on visual control system of acoustic in situ measurement technology for survey of seafloor sediment

Abstract

A new kind of visual acoustic measurement system was developed to obtain the acoustic characteristics of seabed sediment and the working state of in situ measuring equipment in real time. The control unit of the system structure is mainly made up of the underwater control unit and the deck control unit. In comparison with similar traditional systems, the system can transmit the working video image, the measured acoustic data and the waveform synchronous from the underwater control equipment to the deck control unit through the connecting cable. Apart from self-contained in situ measurement, the working modes of it include visual online real-time control. In order to guarantee the stability and measurement precision of the video control system, the process control of in situ measurement was strictly tested in the experimental tank. In 2017, sea trials of the in situ measurement system was carried out in the South China Sea. In the process, data related to measurement, including the acoustic velocity and attenuation coefficient of the seabed sediment, were obtained as well. The results from sea trials show that the control process of the visual system is intuitive, reliable and accurate, and therefore it can be popularized and applied.     

Mediterranean Marine Reserves: Fish Indices as Tests of Protection Efficiency

Abstract. The fish assemblages of a strictly protected area and of a neighbouring fished area, located off a Mediterranean urbanized coast (Carry-le-Rouet, France), were compared in order to select the most evident, constant, and easily recorded indices of the 'reserve effect'. Visual censusing was repeated simultaneously at both sites eight time a year during 3 years along four permanent transects layed on shallow rocky bottoms (9–14 m). Overall species richness was 16% higher in the reserve, but differences in average instantaneous richness were only marginally significant. The most contrasting results were shown by the occurrence frequency, abundance, and demographic structure of two types of target species. The type A target species group comprises 16 large meso- and macrocarnivores particularly threatened by spearfishing, including the common nectobenthic sparids Diplodus spp. Type B target species are represented by two small territorial fishes particularly impacted by angling: the serranid Serranus cabrilla and the terminal phase males of the labrid Coris julis . The size structure of Diplodus spp., S. cabrilla , and C.julis subpopulations strongly argues for the role of protected areas as a reservoir of large spawners. Fishing pressure is assumed to modify the social conditions of the C. julis local populations and to induce earlier sex change.     

GeoVisuals: a visual analytics approach to leverage the potential of spatial videos and associated geonarratives

ABSTRACT

Videos embedded with spatial coordinates, especially when combined with additional expert insights, offer the potential to acquire fine-scale multi-time period contextualized data for a variety of different environments. However, while these geospatial multimedia (GSMM) data include abundant spatiotemporal, semantic and visual information, the means to fully leverage their potential using a suite of visual and interactive analysis techniques and tools has thus far been lacking. In this paper, we address this gap by first identifying the types of tasks required of GSMM data, and then presenting a solution platform. This GeoVisuals system utilizes a visual analysis approach built on semantic data points that can be integrated spatially, which in turn enables management in a unified database with combined spatio-temporal and text querying. A set of visualization functions are integrated in two investigation modes: geo-video analysis and geo-location analysis.     

卷积神经网络和视觉注意力语义分割模型在高分辨率遥感影像分类中的性能分析

随着深度学习语义分割的快速发展,基于计算机视觉语义分割模型的高分辨率遥感影像分类方法也大量涌现。为系统定量地研究经典的和先进的视觉语义分割模型在遥感影像分类中的性能,在总结深度学习语义分割进展的基础上,选择9种基于卷积神经网络(CNN)和视觉注意力的语义分割算法,对米级和厘米级2个尺度的遥感数据集进行分析研究。在模型构建上基于计算机视觉通用的语义分割框架,训练时采用红绿蓝3波段遥感图像并基于ImageNet预训练权重进行迁移学习训练。研究结果表明：通用的语义分割模型通过常规训练设置进行训练能取得较好的遥感影像分类效果,部分地物的交并比(IoU)可以达到90%以上;基于视觉注意力的遥感影像分类模型的精度普遍高于基于CNN的模型,且MaskFormer能更有效地提取离散的地物信息;不同类别的精度最高值并不全在总体最优模型中,部分会存在于次优模型中;类似的地物在更高分辨率遥感数据集中可以获得更高的精度。     

影响云南地区适航条件的视程障碍天气气候特征

利用云南省1981—2013年125站逐日逐6h地面观测的天气现象、能见度和气象要素资料,结合考虑地形高度的高分辨率气象要素场插值技术和不适航雾(能见度≤800m的大雾天气)气象要素判识模型,开展了影响云南适航条件视程障碍天气的气候特征分析。结果表明:云南最主要的视程障碍天气是雾与轻雾,影响适航条件的视程障碍天气仅为雾。不适航雾生成与相对湿度、风速以及地气温差关系密切,"高湿—弱风—稳定"的气象条件是其形成维持的关键,其量化判识特征为:相对湿度98%、风速3m/s、地气温差3℃。滇西南、滇南、滇东是云南不适航雾的多发区,尤其景东—思茅—版纳、西盟—澜沧、保山—腾冲和罗平等地区。滇中、滇西和滇东北是少发区。云南不适航雾日变化呈现早晨和深夜多,午间和傍晚少的特征;季节变化具有夏秋季多于冬春季的特点。与台站观测相比,高分辨率资料客观判识的不适航雾空间分布与地形有合理的对应关系,能够合理地反映地形复杂地区的空间分布特征。未来技术发展可利用卫星反演资料与台站资料同化订正获取高分辨率气象要素资料,利用深度学习等建立不适航雾的判识模型。     

1996—2013年上海虹桥机场低能见度及低跑道视程特征分析

利用1996—2013年上海虹桥机场逐时地面观测资料,对影响机场正常运行的主导能见度低于800 m（低能见度）和跑道视程（RVR）低于550 m（低RVR）时次的出现次数进行统计分析,得到以下结论：1996—2013年虹桥机场除2002年低RVR时次出现较多外,其余年份均为低能见度时次出现更多;低能见度及低RVR时次出现次数分别以57次/年和26次/年的速率减少;季节变化特征显示两者均在12月出现最多,9月出现最少,11月至次年2月是两者出现的高频季节,5—9月为低频季节;一天中低能见度及低RVR多集中出现于19:00—01:00（世界时）,其中23:00达各自峰值;进入夜间后低能见度时次的出现比率首先较大,后半夜开始低RVR影响逐渐凸显。虹桥机场低能见度时次出现次数随能见度数值的降低呈先减少后增多趋势,低RVR时次则在150~200 m及0~50 m两个区间出现次数较多。     

机场低能见度自动观测设备测量数据对比

基于民航机场的实际应用,对比安装在陕西省西安咸阳国际机场跑道一端的大气透射仪和前向散射仪2013年1月—2014年8月测量的跑道视程数据,结果表明:整体上,当跑道视程R > 400 m时,前向散射仪测量数据大于大气透射仪数据的比例高;当跑道视程R≤400 m时,大气透射仪测量数据大于等于前向散射仪数据的比例更高。当跑道视程R≤600 m时,两种设备测量数据具有很好的相互替代性;当跑道视程R > 1000 m时,两种设备测量数据差距较大。在雾、冻雾、霾和暴雨天气时,两种设备测量数据在时间上具有很好的一致性,且在跑道视程R≤1000 m时,两种设备测量数据具有很好的相互替代性,而在雪、烟和扬沙天气时两种设备测量的数据差距较大。     

多种探空仪数据处理显示软件设计

探空业务可获取30 km高度范围内的温湿压风等气象要素.气象科学研究和高空大气实验中使用的探空仪及探空系统有多种,其数据格式也各不相同,为便于使用和分析数据,需要一套通用的处理显示软件.系统采用可视化编程工具Visual C++进行软件系统的设计,实现多种探空仪数据的统一格式转换,它基于MFC单文档视图结构,利用绘图函数实现探空气象要素的垂直廓线显示,并建立了基于SQL Server 2008的数据库,通过编程实现了对探空数据的存储和查询等功能.     

面向飞行器可视导航的复杂空地环境动态3维表示模型

针对飞行器可视导航中复杂空地环境时空要素统一精准表示的难题,本文提出了面向飞行器可视导航的复杂空地环境动态3维表示的GIS数据模型和相应的时刻数据存储结构,并用UML图进行了描述。该模型综合考虑了复杂空地环境要素多维、动态、边界模糊等特点,刻画了多维动态时空数据在空间、时间、尺度、语义四个方面的特征及其相互关系。最后,以某机场的空地环境数据的组织、分析和可视化为例,验证了该模型的实用性。