FEATURE HYPERSPECTRAL IMAGING
The hyperspectral view from space
Abigail Williams speaks to scientists tracking marine plastic using satellite spectral imagery
H
yperspectral imaging and remote sensing are emerging as key tools for tracking plastic pollution in the world’s oceans. There’s
a lot of scientific research using satellite images, and the European Space Agency is supporting a number of remote sensing projects in this area. Projects like Trace, which uses
hyperspectral data from the Italian Space Agency’s Prisma satellite (see the glossary for a list of satellites mentioned in this piece), along with other satellite imagery, to develop a system to track large marine litter and accumulation zones in oceans. Mathias Bochow, environmental scientist
at the Helmholtz Centre, Potsdam, part of the GFZ German Research Centre for Geosciences, is working on the project. He said: ‘While high-resolution, multispectral data enables tracking of floating objects, only hyperspectral data enables material identification. We need to capture a hyperspectral image of a location that allows us to assume where floating litter will be over the next few days to be able to say it is really marine litter and not drifting algae or wood.’ In addition to the Prisma images – and EnMap images once launched – data from multispectral satellites in the PlanetScope constellation were used, alongside the Tiresias oceanographic forecasting model developed by CNR-ISMAR. Bochow noted the ability to track floating objects has only become possible following the launch of the PlanetScope satellite fleet, which captures images from locations around the Earth every day. Although the fleet is typically only used to obtain images over land and coastal zones, images
24 Electro Optics May 2022
were recorded over the Adriatic Sea for this project. ‘High frequent imaging over time enables
tracking of floating objects when linked with an oceanographic forecasting system. We are about to find out how well this works over the next month,’ said Bochow. Spectral capabilities in the shortwave
infrared (SWIR), between around 1,000nm and 2,500nm, are important for identifying plastics, because of the material’s spectral fingerprint at these wavelengths. According to András Jung, co-founder of spectral camera firm Cubert, hyperspectral imaging can be used to sort plastic at a macroscopic level, such as everyday waste products from household or industry. He said there are many successful applications between 400nm and 1,000nm for this. But hyperspectral imaging can also be used to detect microplastics in water, and in this case the SWIR region is more useful. Professor Jonathan Chan, ETRO guest professor at the Vrije Universiteit Brussel, noted that the ideal sensor
‘The potential is not yet fully exploited; scientific communities are just beginning to understand all the pre-processing protocols’
to detect marine plastic should possess more spectral measurement capabilities at SWIR wavelengths. Chan is working on the Muss2 project, which is using spectral and spatial enhancement methods to generate simulated Earth orbit hyperspectral shortwave infrared images and data from the Copernicus Sentinel 2 satellite using spectral response function modelling. Hyperspectral images taken from Earth
orbit are not always available and their coverage is not as large as conventional missions, such as Landsat and Sentinel. To overcome these limitations, Chan said the Muss2 team will apply a sparse theory- based method to enhance multispectral images from the Sentinel 2 satellite. The
expected results are what he described as synthetic Sentinel 2 hyperspectral images at a spatial resolution of 10m, with the same coverage as Sentinel 2 multispectral images. ‘So far, we have been able to generate such images based on Hyperion and Prisma spectral configurations, [and] quantitative assessments are promising,’ he said. ‘In addition, we apply a deep learning-based method for the spatial enhancement of Prisma images to assist in the detection of smaller objects.’ Chan added: ‘All the data inputs
for the project are from open sources, including Sentinel 2, Hyperion and Prisma. The launch of Prisma in 2019 was a major milestone, and soon EnMap will be launched. These are full stretch hyperspectral images at 400-2,500nm with hundreds of spectral bands and represent the most powerful Earth orbit data source for environmental monitoring. The potential is not yet fully understood nor exploited, and scientific communities are just beginning to understand all the pre-processing protocols and data quality issues.’
Signed, sealed, delivered Another interesting initiative is the Spots project, which aims to study the hyperspectral signature of floating plastic debris under less-than-ideal circumstances. As Robin de Vries, geospatial analyst at The Ocean Cleanup non-profit organisation, explained, hyperspectral satellite data has the potential to be a key component in the remote sensing of floating plastic debris. He said: ‘The hyperspectral signature is
affected when plastic is submerged, or when it becomes covered by biological growth over time, known as biofouling. The Spots project is important because many studies have only investigated the detection of floating plastic under ideal conditions.’ The key objectives of the Spots project
are to gather a spectral library of different plastics under varying water depth and biofouling scenarios, and to explore predictive models to estimate depth of water, material type and degree of biofouling from the hyperspectral signature. According to de Vries, the project team has already collected most of the laboratory
@electrooptics |
www.electrooptics.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
