Collaborative Research: Ultraviolet(UV)-MultiSpectral-Polarization 3D Imaging of the Underwater World

合作研究：水下世界的紫外线 (UV) 多光谱偏振 3D 成像

• 批准号: 1636028
• 负责人: Viktor Gruev
• 金额: $ 40.43万
• 依托单位: Washington University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-09-15 至 2017-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1636028&HistoricalAwards=false
• 关键词: Collaborative; Research; Ultraviolet; UV; MultiSpectral

Fine-scale mapping of the underwater world is currently elusive because of a fundamental property of aquatic environments--they are in constant motion. Three-Dimensional mapping of the underwater world in an ecologically relevant way requires mapping not only the physical limits of a specific arena but also the biology within it. Here, the researchers propose to revolutionize the way scientists build near-scale (5-10m) underwater maps by the construction of a UV-Multispectral-Polarization imager with complete multilevel imaging features enabling 3D mapping and full optical characterization of underwater environments. The proposed 3D imager will overcome the challenge of a moving and scattering medium; overcome the problems that cripple conventional scanning devices (e.g. co-registration); while simultaneously filling in the 3D map with biologically meaningful information with images and complete characterization of the light field. With such a device, one will have the capability to map the physical footprint of the underwater world, but also extract species identification from optical characteristics, movement characteristics of organisms within it, health/condition status of biological organisms (e.g. coral reefs, oil spills, plastic contaminants), and comprehensive optical characterization. In addition to providing fine scale mapping of underwater worlds that will serve both biological and conservation missions, the researchers will also use this technology to engage STEM programs in both the Austin and St. Louis areas.This is a Collaborative OTIC award to develop a state-of-the-art 3D imaging device whose purpose is to transform the way researchers map underwater environments and biologically characterize the features within it. The principle investigators propose to develop a high spatial and temporal resolution multispectral polarimeter capable of measuring polarization information in RGB bandwidths combined with three separate and distinct narrow spectral bandwidth channels, one of which being in the UV spectrum. This will produce 12 distinct optical channels that are inherently co-registered, with polarization detection allowing for dehazing capabilities to greatly increase the effectiveness of visual simultaneous localization and mapping algorithms (VSLAM) for obtaining 3D map reconstruction. The co-registered channels will overlay maps with optical information for identifying and measuring benthic characteristics. This next generation underwater mapping device will provide scientists with simultaneous information on (i) physical dimensional space (3D depth), (ii) surface characteristics that identify benthos and organisms within the environment (imaging), (iii) optical characterization of the water column and benthos, as well as (iv) allow for fine-scale tracking of organisms within these underwater environments. This device will enable broad ranges of research questions from oceanographers and marine scientists interested in monitoring coral reefs, animal behaviorists studying 3D camouflage and communication properties, to conservation scientists interested in monitoring environmental degradation (oil and plastic contaminants). This collaborative effort will (a) produce a polarization imaging sensor that captures multispectral polarization information in real-time (~20fps), with low power dissipation and with high spatial resolution, (b) provide dynamic multispectral information on underwater features that were previously unattainable due to scanning technologies with low temporal resolution (~1min), (c) develop software to map and track underwater environments modifying currently developed open source VSLAM software, and (d) test emerging biological hypotheses on camouflage, communication and coral reef monitoring.

由于水生环境的一个基本特性——它们处于不断的运动中，因此目前难以对水下世界进行精确的测绘。以一种与生态学相关的方式绘制水下世界的三维地图，不仅需要绘制特定区域的物理界限，还需要绘制其中的生物学。在这里，研究人员建议通过构建具有完整多级成像功能的紫外多光谱偏振成像仪来彻底改变科学家构建近比尺（5-10米）水下地图的方式，从而实现水下环境的3D映射和全光学表征。所提出的三维成像仪将克服移动和散射介质的挑战；克服阻碍传统扫描设备的问题（例如共同注册）；同时在3D地图中填充具有生物学意义的信息，包括图像和光场的完整特征。有了这样的设备，人们将有能力绘制水下世界的物理足迹，但也可以从光学特征、生物在其中的运动特征、生物有机体的健康/状态状态（例如珊瑚礁、石油泄漏、塑料污染物）和综合光学特征中提取物种识别。除了为生物和保护任务提供精细的水下世界地图外，研究人员还将利用这项技术参与奥斯汀和圣路易斯地区的STEM项目。这是一项合作项目，旨在开发一种最先进的3D成像设备，其目的是改变研究人员绘制水下环境的方式，并对其中的特征进行生物学表征。主要研究人员提出了一种高时空分辨率的多光谱偏振计，能够在RGB带宽下测量偏振信息，并结合三个独立的窄光谱带宽通道，其中一个是在紫外光谱中。这将产生12个不同的光学通道，这些通道本质上是共同注册的，偏振检测允许除雾能力，从而大大提高视觉同步定位和映射算法（VSLAM）的有效性，以获得3D地图重建。共同注册的通道将覆盖带有光学信息的地图，以识别和测量底栖生物的特征。下一代水下测绘设备将为科学家提供以下方面的同步信息：(1)物理维度空间（3D深度），(2)识别环境中底栖生物和生物的表面特征（成像），(3)水柱和底栖生物的光学特征，以及(4)允许在这些水下环境中对生物进行精细跟踪。该设备将使海洋学家和海洋科学家对监测珊瑚礁感兴趣，动物行为学家研究3D伪装和通信特性，保护科学家对监测环境退化（石油和塑料污染物）感兴趣。这项合作将(a)生产一种偏振成像传感器，可以实时捕获多光谱偏振信息（~20fps），具有低功耗和高空间分辨率；(b)提供水下特征的动态多光谱信息，这是以前由于低时间分辨率（~1min）的扫描技术无法实现的；(c)开发软件来绘制和跟踪水下环境，修改目前开发的开源VSLAM软件。(d)测试关于伪装、通讯和珊瑚礁监测的新兴生物学假说。