Sea Technology

SEP 2017

The industry's recognized authority for design, engineering and application of equipment and services in the global ocean community

Issue link: http://sea-technology.epubxp.com/i/875429

Contents of this Issue

Navigation

Page 32 of 68

32 st / September 2017 www.sea-technology.com By processing 48,000 images on a single pile, it was possible to gener- ate a high-density and high-accu- racy cloud of about 14 million 3D points. The space between points (resolution) was less than 1 mm. Economic Considerations The economic impact of the VLS technology is significant for techni- cal, operational and financial rea- sons. First of all, the data capture is very fast and reduces considerably the vessel time. Video shooting takes only minutes and is often al- ready planned for visual inspection purposes. A typical spool metrolo- gy survey would take no more than 30 minutes. Second, nonspecialists can car- ry out data capture after a short training on basic rules of photogrammetry and DimEye technologies. This means less people needed on the boat and less people involved in op- erations. Third, the equipment is off-the-shelf, and HD cameras already integrated can be used without any problem. They can also very easily be rented from traditional suppliers of offshore equipment. There is no need for very expensive and sophisticated sensors that could only be used by high-level specialists. Finally, data processing is becoming faster and faster thanks to the development of new high-performance cam- eras and the development of powerful image-processing algorithms. New Developments The future of 3D measurement by optical techniques such as photogrammetry and video laser scan will be driven by the development of the hardware and new methodolo- gies associated with software development. With the hardware being now mostly off the shelf, the technology will automatically benefit from any camera im- provement. Video cameras now have 4K resolution and will have 6K or more in the future. Stabilization techniques will also help by allowing faster ROV speeds. Ongoing research on high dynamic range (HDR) video will certainly lead to the continuous improvement of images. Software development will be driven by the market and customer requirements. Although data processing has be- come significantly faster over the last few years, there is still a demand for shorter time frames, including real-time re- sults. This will require more automation through the devel- opment of more dedicated software, such as applications to be used by nonspecialists. ST Arnauld Dumont is a surveyor-expert engineer and has been working for more than 25 years in the field of 3D measurement by photogrammetry and laser technologies in various industries (oil and gas, nuclear, aerospace, civil engi- neering). He managed 3D survey projects in the U.S., Europe, Africa and Asia and founded DimEye Corp. in Los Angeles back in 2009 with the goal of focus- ing on applications for inaccessible environments such as subsea, space and nuclear areas. Since then, DimEye has developed the innovative technology called VLS for subsea and aerial periodical inspection of infrastructure compo- nents and 3D as-built surveys. sions. These measurements can also be sent for finite element analysis (FEA) if necessary. Mooring Chain. Measuring moor- ing chains is a difficult challenge. In order to achieve 360° of coverage, multiple ROV runs must be done from both "sides" of the chain. This process becomes complicated when environmental factors such as ocean currents and chain movements are taken into account. Conditions like this make ROV maneuvers intense. However, thanks to its high level of flexibility, the VLS technology makes it possible to capture HD video, gen- erate accurate point clouds, measure general dimensions and determine the grip zone wear of each chain link to a submillimeter level of accuracy. Piles. In 2016, DimEye was contracted to measure the interior of subsea piles in the Gulf of Mexico. A specific VLS sensor configuration was established in order to completely cover the internal surface and generate an accurate point cloud from which 3D analysis was carried out (such as cy- lindricity and ovality), as well as collision analysis. With a total of 768,000 images—48,000 per pile—this is the largest close-range photogrammetry project ever realized. (Before this, the largest was the 3D survey of the nuclear fuel repro- cessing plant of La Hague, France, which produced close to 200,000 stereoscopic couples—a total of 400,000 images.) Zoom on a sample 3D mesh of subsea piles showing density and accuracy.

Articles in this issue

Links on this page

Archives of this issue

view archives of Sea Technology - SEP 2017