Designed To Meet Demanding Delivery Targets
With the increasing number of satellites in the DMC constellation and their greatly increased data collection capacities, there is a need for an ever more rapid and accurate assessment of image data quality to meet stringent delivery times of our customers.
DMCii has addressed the issue by developing a unique suite of data quality measures to provide automated data quality assessments and in some cases automatic correction of imagery products.
Data quality is a fundamental element along with radiometric calibration that defines the final product quality for the end-user. With large data volumes it was often impossible in the past to guarantee the data quality of every individual image.
DMCii has therefore developed the necessary methods to not only reduce the occurrence of data quality issues, but also predict and prevent them in many cases using proprietary algorithms.
The advantages of these automated methods include:
• Very high quality, artefact free images.
• Identification of potential problems in imagery before they impact the visual quality of the data.
• Reduction in system down time for re-calibration of the sensor, hence greater availability of the satellite systems for customer acquisitions.
• Minimum operator interaction, so the operations team has more time for customer requests.
• Automatic notification of image anomalies.
Technical issues evaluated include individual detector anomalies, calibration drift, changes in system Signal to Noise Ratio, charge particle impacts (zipper effects), striping artefacts and saturation specular effects.
Detector drift although still not visible in the imagery (even with an extreme stretch) can be detected as a spike in the plot of the image statistics.
The image on the left shows a typical anomaly from single detector drift (not visible, even under an extreme stretch), we magnified the anomaly on the image on the right to show its location, based on the image statistics shown in the diagram above.
This is the first system which evaluates its performance using normal image acquisitions, allowing long time series of image performance to be generated. This is one strand of our data quality procedures.
A second strand is our support for the CEOS initiative QA4EO (Quality Assurance for Earth Observation). This initiative has been supported by DMCii since its inception in 2007. The aim is to provide Quality Indicators related to the uncertainty in any delivered product. DMCii has already taken the first steps and has a development plan to achieve QA4EO compliancy.
As part of the services portfolio, DMCii provides both guidance and direct support in image evaluation for satellite operators and end-users wishing to understand artefacts or characteristics of particular sensors or data sets. Contact firstname.lastname@example.org
DMCii provides calibration services for all member satellites of the constellation.
The two key elements to achieve high quality imaging data are accurate, stable calibration and automated efficient data quality. DMCii have been calibrating satellites for over 10 years and effectively link all the satellites in the constellation to a “gold” standard, that of Landsat 7 using advanced methods of cross-calibration.
Very high accuracy of temporal correlation of Landsat 7 data and UK-DMC-2 data to give a relative uncertainty in calibration of less than 1%.
The methods used are proprietary and have placed DMCii at the leading edge of developments in the field, presenting the advanced methods at the internationally recognised CEOS Working Group on Calibration and Validation (WGCV) of which it is the only commercial associate member and its subgroup the Infrared and Visible Optical Sensors (IVOS).
DMCii has led the way in development of robust cross-calibration methods over the Dome-C site and Dr. Mackin was the lead author on a European Space Agency Study in collaboration with the National Physical Laboratory defining the effective procedures that need to be applied and adopted, these methods are endorsed by the CEOS IVOS subgroup
The Dome-C station in Antarctica is a key site for cross-calibration © NASRDA
DMCii has also taken a leading role in trying to understand another key calibration site, the Libya 4 site. This is a large dune-field in a desert region of North Africa that has shown remarkable stability for several decades and hence provides a suitable target for long term cross-calibration. This site has been the key site for cross-calibration against Landsat 7.
The complex dune field provides a challenge to high accuracy calibration, solved by a novel set of measurement and model procedures developed by DMCii.