![]() ![]() ![]() This is about 1/10th of all hard disk storage produced in 2017 and does not consider repeated monitoring for time series observations. An uncompressed, color ortho-photo of the entire seafloor, acquired at 1px/mm resolution, would require ca. The scale of the image data management challenge is governed by the required image resolution and the area to be surveyed. The trend towards parallel deployment of multiple AUVs will further increase the pressure in being able to efficiently curate and manage those big image data sets. AUVs are being deployed for large-scale assessments of the seafloor which require specific data processing workflows 5. These can record large volumes of image data at an unprecedented acquisition velocity. Together this often leads to un-managed data in the form of dispersed copies on mobile hard disks which unnecessarily duplicate the data, prevent access controls and easily get lost or corrupted.Īn additional need for more standardization exists due to the increasing popularity of autonomous underwater vehicles (AUVs). Subsequent steps like manual image annotation and automated image analysis are even less standardized. Currently though, protocols are lacking for the steps following the marine image acquisition, namely these are: i) image data curation to quality control the recorded raw data and ii) image data management to publish the data sets in a sustainable way in work repositories and long-term data archives. These include the design and deployment of underwater camera gear for scientific and industrial applications 1, the curation and management of oceanographic data 2, the acquisition of all data required for a full biological assessment of a habitat 3 and references for manually annotating marine imagery 4. A multitude of strategies have been developed for various marine data acquisition and data management aspects. These record valuable data for navigation, exploration and monitoring purposes. Modern ocean science gear for underwater sampling is commonly equipped with optical imaging devices like photo and video cameras. We describe guidelines for data acquisition, curation and management and apply it to the use case of a multi-terabyte deep-sea data set acquired by an autonomous underwater vehicle. Here, we present a workflow towards sustainable marine image analysis. Hence, the expensive data acquisition must be documented, data should be curated as soon as possible, backed up and made publicly available. Systematic data analysis benefits from calibrated, geo-referenced data with clear metadata description, particularly for machine vision and machine learning. ![]() This generates a need for automated data processing to harvest maximum information. Volume and velocity are further increased by growing fleets and emerging swarms of autonomous vehicles creating big data sets in parallel. Technological advances like 4K cameras, autonomous robots, high-capacity batteries and LED lighting now allow systematic optical monitoring at large spatial scale and shorter time but with increased data volume and velocity. Diving robots, towed cameras, drop-cameras and TV-guided sampling gear: all produce image data of the underwater environment. Optical imaging is a common technique in ocean research.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |