James is a Channel Manager EMEA at Exelis VIS; he studied Geographical Information Systems at Kingston University London. With 10 years experience in geospatial market in sales/consulting/management, James is an advocate who loves to help teams and end users benefit from geospatial approaches. Enjoy music, running, cycling, surfing (I try!) and snowboarding.
Author: James Slater
In Greece, the forest fire season starts the first of May each year and ends October 31. During this period which coincides with the tourist season, the civil protection mechanism and fire services are in a high alert status and operate under a well-planned scheme. However, local administration authorities and municipalities do not yet provide access to mapped assessments for burned areas immediately after or during the occurrence of a wildfire event and instead rely on central government agencies and institutions through what can be a time-consuming bureaucratic and fiscal process.
For the last three years as part of Inforest’s CSR initiative and bundled with an environmental remote sensing outreach initiative for local government, Inforest Research has made efforts to extend their knowledge and experience in the application of remote sensing technologies to deliver rapid forest fire assessment damage maps. By relying on free remote sensing data sources (MODIS and more recently Landsat 8 OLI datasets) in combination with semi-automatic processing in ENVI, the resulting map products for burned areas and major fires occurrences are then made freely available for institutions to use.
Chios & Rhodes Island Analysis
From these assessment maps, Inforest Research additionally produces newsletters distributed directly to local authorities and makes available the methodology to enable institutions and users to repeat and implement the process. Newsletters, written in Greek are available to download, while a video explaining the concept and methodology can be viewed on Inforest’s YouTube Channel. Beneficiaries of this work include not only local government but also conservation NGO’s such as WWF Greece.
Examples of the Inforest Research Forest Fire Damage Assessment newsletters for events in 2013 and 2014.
Inforest Research hope these efforts will help to improve the understanding of and response to the impact of forest fire events whilst improving local agency capabilities to conduct this analysis in house through sharing of methodologies, implementation and outcomes.
Categories: ENVI Blog | Imagery Speaks
Recent developments in remote sensing are leading to rapid change posing unique challenges and opportunities to users and providers of related technologies. So what is the transformational change? To answer this lets first look back at Earth’s Satellite History to get a perspective. Take a look at this animation of metadata trails http://vimeo.com/90127911 captured from all active Earth Observation Satellites 1986-2012 and see what happens around 2007… the period 2007 to 2012 brought a 25% increase in active earth observation missions leading to a much greater availability of data about the earth than at any other time.
New sensors contributing to this change include WorldView-3; expected to launch mid-August 2014 promises to be the first very high spatial resolution, multi-payload, super-spectral, commercial satellite offering revisit times of less than a day. In recent weeks DigitalGlobe also announced that with the U.S. Department of Commerce relaxing restrictions their highest resolution imagery: 0.25m panchromatic and 1.0m multispectral will soon be available.
Radar Imaging or SAR continues to show incredible operational capabilities from monitoring sea ice, oil spills, land-use change, land deformation and disaster response and April this year saw the launch of a new SAR Satellite: Sentinel 1-A. Part of the European Copernicus project the really great thing here is that Sentinel 1-A promises free and open access removing a significant barrier to entry. SAR is really an accelerating area... unlike imagery which observes sunlight reflected from the surface, SAR is an active instrument which emits a pulse of millimeter wavelength light so SAR sees though cloud cover and it is available 24 hours, day and night and because SAR is active it can identify small scale features such as changes in surface height and texture. SAR’s unique capabilities of cloud penetration, night illumination, and water height detection make it a perfect fit for proactive monitoring of pre, event and post event flood monitoring.
New approaches to data acquisition are also leading to new possibilities… constellations of cube satellites developed from COTS hardware re-purposed from consumer applications are set to dramatically increase the amount, timeliness and availability of data about of our world. SkySat from SkyBox is such an example driving access to up to date, relevant and timely data.
Other sources of imagery are also coming to the fore. Previously the domain of military users Unmanned Aerial Systems (UAS) and Full Motion Video (FMV) are starting to deliver hyper localized and real time information from the earth’s surface.
Away from sensors and satellites the way information about the earth is consumed has also changed, the realization of image services facilitating live-stream analysis of data breaking the bond between user and data location and creating new possibilities for creating derived information without the overhead of data ownership.
The result of these ground breaking developments is a data avalanche. This data needs to manipulated, converted, integrated and analyzed to extract meaningful information. So How Does Exelis Help? We convert the 1s and 0s coming from the full range of data types spanning optical, SAR and LiDAR into meaningful, actionable information. Our tools workflow automation, interoperability and new approaches to distributed, networked processing meet the challenge of this rapid increase in data availability and with large scale projects such as Copernicus in Europe and others around the world the trend towards improved availability, quality and timeliness of data is set to continue.
In this short blog post I have no doubt omitted developments you consider key. Let us know what you’re looking forward to in the comments below.
DigitalGlobe is a big player in the provision of satellite imagery, so it’s no surprise that the earth observation community is looking forward to the launch of their new satellite, WorldView-3. The company already operates a constellation of satellites that are widely used in commercial image analysis (QuickBird, WorldView-1 and WorldView-2, http://www.digitalglobe.com/about-us/content-collection#overview). But besides the obvious benefits of more satellites giving greater coverage of the Earth’s surface, what other benefits will WorldView-3 be bringing to the party?
Well first, a bit of background. Like its sister satellites QuickBird and WorldView-2, WorldView-3 (WV-3) is multispectral. This means that it is observing light reflected from the Earth at a number of different frequency bands across the electromagnetic spectrum (unlike WV-1 which only observes one bands of frequencies, i.e. panchromatic). Typically these bands are sited within the red, green and blue portions of the spectrum, with a fourth band within the Near Infra-Red (NIR). The three RGB bands allow us to produce images as the human eye would see, whilst the NIR band is useful for delineating features such as land/sea boundaries or identifying clouds -- basically any feature where we’d expect a sudden change in temperature. This is the approach of QuickBird and many other commercial satellites.
WorldView-2 takes this approach a little further, by utilising 8 bands in the Visible and NearInfra-Red spectra (VNIR). I’ve listed the bands used by all of these satellites in the accompanying table. We can see that as well as the standard RGBand NIR, WV-2 has some additional bands -- Coastal Blue, Yellow, Red Edge and a second NIR band. Each of these bands is chosen to be sensitive to a particular feature on the surface. For example, the Coastal Blue band is more sensitive to moisture within an image than the traditional Blue band, so it can be used to improve on water depth measurements around harbours. These additional bands really help with image analysis because the additional information allows for more accurate image classification and segmentation which would be carried out with analysis software such as ENVI.
WorldView-3 takes the advances made with WorldView-2 the next step forward and includes 8 more bands within the Short-Wave Infra-Red (SWIR) region of the spectrum (http://www.satimagingcorp.com/satellite-sensors/WorldView3-DS-WV3-Web.pdf), in addition to the VNIR channels of WorldView-2 (also listed in the table). This allows WV-3 to observe a much wider range of the electromagnetic spectra than most other commercial satellites, and will allow us to start looking for the individual spectral signatures of materials. The accompanying plot shows the spectral signatures of three minerals. The light blue regions on the plot denote the VNIR bands covered by WV-2and WV-3. The light red regions denote the SWIR bands covered by the 8 new WV-3 bands. This shows just how much further the new bands penetrate into the electromagnetic spectrum. The spectral profiles of each pixel in an image can be compared to a spectral library (such as the reflectance spectras shown) to classify what material is contained within that pixel.
This type of automated spectral classification is commonly carried out in ENVI with hyperspectral data (from satellites containing hundreds of spectral bands) which can sometimes be costly to acquire. However, the ability to remotely monitor materials is invaluable to a great number of industries for example in forestry applications where we see users wanting to monitor tree health and pest infestation in remote regions. WorldView-3 brings a limited version of this capability to users at a reduced price, so we still won’t expect the accuracy of a hyperspectral satellite, but it really is a step beyond the features we could extract from multispectral data. So, I’ll stick with Digital Globe’s marketing and refer to it as ‘super-spectral’.
WorldView-3 will be launching in the next few months, so the first data is likely to be available at the beginning of 2015. As a partner of Digital Globe, we’ve been preparing for this. In fact, our parent company developed the SWIR sensor aboard http://www.exelisinc.com/news/pressreleases/Pages/Exelis-delivers-first-of-its-kind-commercial-payload-to-DigitalGlobe-for-WorldView-3-satellite.aspx). The data will be able to be loaded into ENVI as soon as it’s available, so I advise you to keep an eye out for further updates because this instrument looks like it will be a great addition.
Tags: multispectral imagery, Satellite Imagery, earth observation, WorldView-3, Digital Globe
If you’re a regular reader of this blog you’ll know about the diverse applications of ENVI in environmental monitoring, change detection and the identification and categorization of features, processes and flora at or close to the earth’s surface; from populating geodatabases with bare earth models from LiDAR point clouds to the use of hyperspectral imagery to assess the chemistry and structure of the worlds forests and even the use of surveillance technologies such as radar to monitor and bring action against oil polluters in our oceans ENVI has proven credentials and flexibility to answer to different types of challenges. One application area that perhaps receives less coverage but is of growing interest is its use in monitoring earths' fauna.
Following the groundbreaking work of Peter Fretwell and the British Antarctic Survey (BAS) where ENVI was used to double current estimates of emperor penguin populations in Antarctica, the BAS scientists have now turned their attention to the study of Southern Right Whale populations in the Golfo Nuevo on the coast of Argentina. The study published in PLOS One answers to the notoriously difficult challenge of estimating whale populations, a task usually undertaken by costly and oftentimes dangerous manual observation methods. The new approach represents the first successful attempt of the use of satellite imagery to count whales and suggests a transferable methodology which could have far reaching implications for future surveys.
In order to spot the whales the team first manually interrogated a WorldView2 VHR image to retrieve a baseline count from which results from an automated detection approach could be evaluated. The WorldView2 image offers nine bands including eight colour bands at 2M resolution and one panchromatic band at 50cm. Band 5 (termed the coastal band) was especially useful for this task as it collects light at wavelengths between 400nm and 450nm in the blue/violet part of the spectrum which can penetrate water at depths of up to 15M increasing the chances of a positive whale identification. For the visual inspection bands 1 (red), 8 (NIR2) and 5 (coastal) proved to be the most useful positively identifying 55 probable whales.
Single band images of a probable right whale in the satellite image from each of the eight multispectral bands and the panchromatic band of the WorldView2 data. Reprinted under a CC BY license with permission from British Antarctic Survey and DigitalGlobe.
Next ENVI 5 was used to automate the process and compare results to the manual identification. Supervised and unsupervised classifications and single band histogram thresholding were trialed with thresholding of the panchromatic and coastal bands delivering the best results confirming 89% of probable whales identified in the manual count.
Comparison between manually identified and automatically identified whales.
From what is already a robust set of results the team at BAS identified several ways the approach can be refined to achieve even better accuracies; the use of multiple sets of images to remove false positives generated from objects such as rocks and boats would be straightforward. Placing reflectors below the sea surface would give better insight into how far the satellite sensor is seeing into the water necessary to formulate a ratio of visible whales versus invisible whales from which population estimates could be usefully drawn. Further this study concentrates on pixel-based analysis, object-based analysis which adds spatial, and textural classification techniques such as those employed by ENVI Fx could help to further refine the automated process.
Results from this study are extremely encouraging and point the way to new automated detection methodologies set to revolutionize scientists ability to monitor global whale population dynamics with greater accuracy, frequency and at a reduced cost and risk over traditional methods.
Tags: ENVI, Academic, Remote Sensing, Image Analysis, multispectral imagery, environmental monitoring
I want to talk about the weather, and not only because it’s a national pastime for us Brits (and so nicely fits the stereotype) but because it’s getting weird.
2013 saw some of the most extreme weather experienced in decades. It was the 7th hottest year on record and featured one of the strongest cyclones ever recorded. Some of the longest and most prolonged heat waves culminated in Australia’s hottest ever year (fully 12 of the 14 hottest years on record are from this millennium). In the UK we experienced the wettest ever recorded summer in 2012, followed record lows in March 2013 and then series of high intensity gales coupled to storm surge tides resulting in widespread flooding towards the end of 2013 and into 2014. Our headquarters in Boulder, Colorado saw an average year of rain fall within just one week in September 2013 causing deaths, mass evacuations and the loss of hundreds of homes and businesses, and as we moved into 2014, Midwest, North-East and Southern US states were gripped by cold weather reaching -30C in places. The cumulative effect of an estimated 1.5 million weather related accidents each year in the US alone results in approximately 700,000 injuries, 7000 deaths and $42B in economic costs.
The trend for increased change is set to continue. Towards the end of 2013 scientists reported how new climate models taking greater account of cloud changes indicate heating will be at the higher end of expectations with predictions that the planet is likely to warm by 4C by 2100 resulting in greater climate instability and increased exceptional weather events. The debate around the anthropogenic impact on climate change continues and how effectively a causal link can be established between this and the frequency of extreme weather continues to be an area of intense investigation. Where all parties can agree is the need to promote improved understanding, monitoring, assessment and response to the effects of extreme weather events. Earth Observation and Remote Sensing play a key role in mitigating social, economic and environmental impacts of linked effects and here ENVI and IDL have a long and successful history.
Exelis are proud to deliver solutions which help scientists study climate trends and forecast weather conditions through extensive system, sensor and analytical support as well as driving awareness and understanding. Examples include The Global Precipitation Measurement (GPM) mission, where IDL underpins radar systems providing next-generation global observations of rain and snow, further examples include Solar wind visualization at NOAA and Storm Tracking for the Applied Physics Laboratory. In other areas ENVI delivers solutions to facilitate the passage of ships through sea ice, map wildfire extent and facilitate flood detection. Most recently Exelis innovated again delivering Helios, a new approach to the challenge of identifying and monitoring severe and threatening weather at a hyperlocal scale by applying Exelis image science to an aggregated network of thousands of cameras across the United States.
As we understand more about our changing environment demand for robust tools to meet the needs of both the scientific community and specific consumers of weather and climate data continue to grow, and Exelis look forward to future collaborations and developments in this most important field.
Tags: Remote Sensing, weather, climate
Sign Up for News & Updates: Stay informed with the latest news, events, technologies and special offers.