Between the Poles

Urban planning working group announced by the Open Geospatial Consortium

An Urban Planning Domain Working Group (DWG) has been chartered to define the role for Open Geospatial Consortium (OGC) standards within the urban planning discipline.  The DWG offers an open forum for the discussion and presentation of interoperability requirements, use cases, and implementations of OGC standards relevant to urban planning. 

Technologies and trends such as smart cities, smart grids, sensor webs, the Internet of Things (IoT), facilities and asset management, indoor and outdoor navigation, energy performance modeling and real-time, “big data” analytics are important for urban planners.  In these technology domains, open standards encourage the sharing of information. The OGC Urban Planning Domain Working Group intends to discover requirements for open spatial standards in information systems involved in the planning, design, use, maintenance and governance of publicly accessible spaces. Requirements identified by OGC Domain Working Groups are typically used as the basis for standards development by chartered OGC Standards Working Groups (SWGs).

Image: Sunshine Ecomap

OGC's new Urban Planning group releases draft charter

An Urban Planning Domain Working Group (DWG) is being chartered to define the role for OGC standards and related activities related to urban planning.  It is also intended to provide an open forum for the discussion of interoperability requirements, use cases, pilots, and implementations of OGC standards in this domain.  The OGC seeks comments from the public on the draft new charter. The comment period closes on  6 August 2014.

Technologies and trends such as Augmented Reality (AR), Smart Cities, Smart Grids, Sensor Webs, the Internet of Things (IoT), LBS (Location Based Services), Facilities Management, navigation (indoor and outdoor) and “Big Data” Analytics all can play important roles in informing urban planners.  In these technology domains, open standards can facilitate the development, publication, discovery, and use of information. The OGC Urban Planning DWG intends to discover requirements for open geospatial standards in information systems involved in the planning, design, use, maintenance and governance of publicly accessible spaces.

The Internet of Things forecasted to include 26 billion intelligent devices by 2020

In 2009 there were about 10¹⁹ transistors shipped in 2009, or over 1 billion per person.  It is estimated that the  number of smartphones, tablets and PCs in use will reach about 7.3 billion units in 2020.  It is projected that component costs will become so inexpensive that connectivity will become a standard feature even for devices costing less than $1.

According to a Gartner report, the Internet of Things (IoT) which Gartner defines as the "network of physical objects that contain embedded technology to communicate and sense or interact with their internal states or the external environment" and which excludes PCs, tablets and smartphones,  will expand to 26 billion units installed by 2020 up significantly from 0.9 billion in 2009. 

According to Gartner the IoT includes the devices (hardware), embedded software, communications services and information services such as geolocation associated with the things. Gartner refers to the companies that provide the hardware, software and services as IoT suppliers.   Gartner said that IoT product and service suppliers will generate incremental revenue of about $309 billion, mostly in services. 

The economic value-add which represents the aggregate benefits that businesses derive through the sale and usage of IoT will reach $1.9 trillion across all sectors in 2020.  The verticals that are leading the adoption of IoT are manufacturing (15 %), healthcare (15 %) and insurance (11 %).  The IoT includes sector-specific technology such as automated manufacturing systems and more generic technology such as "smart building" technologies.

A recent study by Cisco of the Internet of Things (Cisco calls it the Internet of Everything) estimated that it represents a $19 trillion global opportunity over the next decade.

Standards for the Internet of Things

Geospatial information is a cross-cutting requirement of many intelligent devices participating inthe IoT.  The Open Geospatial Consortium (OGC, which creates and manages standards that enable geospatial interoperability, and the International Telecommunications Union Telecommunication Standardization Sector (ITU-T) are jointly collaborating on the advancement of standards for the Internet of Things (IoT).  As part of this effort the OGC has submitted the OGC Open GeoSMS standard for adoption through the ITU.  In addition the ITU-T has just published a ITU-T Technology Watch Report: Location matters: Spatial standards for the Internet of Things, written by staff and members of the OGC in collaboration with the ITU Secretariat.

Smart cities market in Asia projected to grow rapidly through 2023

The Asia Pacific region is a primary driver of global urban development trends and is where some of the world's the most ambitious attempts to direct those trends to national development goals.  Increasingly, smart city technologies are seen as means to keep metropolitan and national economies competitive, and Asia Pacific companies are among the leaders in smart city solutions.

According to BSRIA, there are many smart city initiatives underway in Asia.

Some examples are Songdo IDB in Korea and Fujisawa in Japan - two smart cities already under development. China has 36 smart cities in development and a low carbon model city in Tianjin as well as 104 nationally selected smart city demonstration projects.  Singapore plans to become a smart nation by 2015.  Iskandar is Malaysia's first smart city.  After a devastating earthquake, Christchurch, New Zealand is being redeveloped as a smart city.

Two of these I am acquainted with because I visited the planning office for the DMIC the last time I was in India and  I met with several people involved with the Songdo project quite a few years ago in Korea.

Delhi Mumbai Industrial Corridor

In India the Delhi Mumbai Industrial Corridor (DMIC)is intended to be a smart city.  The Delhi-Mumbai Industrial Corridor is a huge US$ 90 billion project 1,483 km in length between Delhi and Mumbai.  The objective of the DMIC is to create a base for economic development by providing  access to the best state-of-the-art infrastructure.

Songdo IDB

Songdo IBD is much smaller scale and has very different objectives.  It is a completely new city, being built from scratch on 1,500 acres of reclaimed land along Incheon's waterfront.  Songdo represents a serious attempt to create a sustainable, liveable city.  The city will be the first LEED certified district in Korea and the largest project outside North America to be included in the LEED ND (Neighborhood Development) Pilot Program.

Asian smart city market sizing

Navigant Research has released a report  that provides an assessment of smart city initiatives in Asia Pacific and profiles the key industry players in the region.  The report attempts to size the smart city market through 2023 by region (Australia/New Zealand, Greater China, India, Japan, Southeast Asia, and South Korea) and sector (smart energy, smart transportation, smart water, smart buildings, and smart government).

Navigant points out that the smart city market is a loose collection of several markets that are concerned with addressing urban problems in some way.  Navigant considers the smart city market to be comprized of five industry sectors: smart energy (and the smart grid), smart transportation, smart water, smart buildings, and smart government (including social services such as health, education, and security).

The Navigant report examines smart city drivers, policies, and development prospects in six distinct Asia Pacific regions: Australia/New Zealand, Greater China, including Hong Kong and Taiwan, India, Japan, Southeast Asia, including Indonesia, Malaysia, the Philippines, Thailand, Vietnam, and Singapore; and South Korea.  Navigant estimated that these regions make up most of the addressable Asia Pacific market for smart city technology and include most of the cities that have active smart city development initiatives.

The uderlying trends driving the smart city market in Asia Pacific include continuing economic growth in China, India, and Southeast Asia; a focus on upgrading infrastructure in  China, India, Australia, and South Korea; and high expected levels of smart city technology adoption in Japan, Australia/New Zealand, and South Korea. Navigant estimates that the annual smart city technology investment will almost quadruple to $11.3 billion by 2023. This just represents the smart part of much larger smart city projects: smart sensors and meters, dedicated networks, cloud computing platforms, data analytics, and integrated systems and applications across different industries and sectors.

The smart energy and smart transportation sectors are the largest and most developed.  The smart government sector which includes security, health, and education applications is the fastest growing.  Smart water systems are currently a small part of the overall market, but will experience substantial growth to meet the needs of expanding cities in water-scarce areas such as India, Northern China, and Australia. Smart buildings will become an appreciable part of the overall $ 7 trillion construction market as the benefits of smart city technology become apparent and increasingly stringent energy intensity requirements are mandated throughout the region.

Smart wastewater system enables city to avoid $100 million sewer expansion

McKinsey&Company has published interesting interviews with mayors of U.S. cities on how open data not only makes for more transparent government, but also improves the effectiveness and efficiency of services. Pete Buttigieg, Mayor of South Bend, Indiana, described how his city is using data to improve the efficiency of wasetwater infrastructure that supports its 100,000 citizens.

Pete Buttigieg's perspective is that all of the data belongs to the people because the City created it while working for them. The City's challenge is to make it available—and not just available but intelligible. It’s not transparent if you just dump a giant pile of data online. The City worked very hard to make the data not only accessible but also usable on their open data portal.

Reducing combined sewer overflows

In the U.S. about 770 cities mostly in the Northeast and Northwest have combined sewers. I've blogged previously about EPA consent decrees designed to reduce the discharge of untreated sewage into surface waters with Cleveland, Akron, Pittsburgh, Atlanta, and Chicago.  These have imposed legally binding investments of between half a billion to multi-billions of dollars on these cities, for many of which this represents a huge financial burden.

About two years ago, the EPA annouced an agreement with the City of South Bend which required the City to make an estimated $509.5 million worth of improvements to its combined sewer system to significantly reduce overflows of raw sewage to the St. Joseph River, which among other things is used for kayaking compeititions.

The City of South Bend ws one of the first cities to move its wastewater management into the cloud.  South Bend has invested $6 million to implement the smart sewer system, but they estimate that it has helped avoid $100 million in capital expense by better using the infrastructure they already have.  It has enabled them to stay ahead of their EPA consent decree and avoid the very expensive capital construction projects that other cities have faced.

Implementing smart sewers in the cloud

South Bend's combined sewer system collects storm water, sanitary sewage, and other pollutants which it conveys to South Bend's wastewater treatment plant (WWTP).  During wet weather events, and during some dry weather time periods, a portion of the sewage that flows through the combined sewers is not conveyed all the way to the WWTP; instead the raw sewage is discharged into the St. Joseph River.  Currently, on average South Bend discharges over 2 billion gallons of untreated sewage resulting from 80 combined sewer overflow (CSO) events annually.

South Bend partnered wit the University of Notre Dame and local technology company EmNet to develop a way to keep sewers from overflowing.  In 2005, EmNet created a system called CSOnet (Combined Sewer Overflow Network) which relies on wireless sensors installed under manhole covers to monitor water levels in sewer pipes. The system consists of 115 sensors that communicate live updates every five minutes wirelessly to a central control center.   Based on the water levels measured by the sensors, "smart" valves can be opened or closed to redirect flow  into pipes where capacity is available.

South Bend is using IBM's Intelligent Operations Center for Smarter Cities software-as-a-service. The cloud service allows public works officials to view aggregated data related to water management in real time, to help predict where incidents will occur.

Benefits

in its first year of operation, the cloud based wastewater management system has practically eliminated dry weather wastewater overflows and reduced wet weather overflows by 23%. If they had done what other cities have and simply expanded the sewer system they estimate it would have cost the city an additional $120 million in capital expenditures.

Itron study finds that the utility industry needs to reinvent itself through technology

Itron has published its first annual Itron Resourcefulness Index. It is a study based on a survey of utility executives from the gas, water and electricity industries as well as utility consumers. 800 informed consumers and 600 utility executives in 14 countries were asked to respond to questions about the state for the global utility industry.

Concern about the current state of the utility industry

Respondents indicated that there is a significant level of concern about the state of the utility industry.   60 percent of utility executives and 70 percent of customers believe that utility operations are not efficient. 70 % of utility customers and 40 % of utility executives agreed that operational inefficiency lowers customer satisfaction.  The discrepancy between the utility executives and consumers points to another problem identified in this study.

94 percent of utility executives surveyed agreed that transformation of operations was essential to improve efficiency. The concern was greatest in the developing countries, especially China and India. Developed countries, in particular Spain, Australia and  Canada, had the lowest overall concern among both consumers and executives.

Noteworthy is the gap between the level of concern expressed by utility executives and consumers.  In every country consumers expressed greater concern about the state of the utility industry than utility executives.

Eight out of 10 consumers said they do not receive enough information from their utilities today, and seven out of 10 said they want to know more.

The survey also asked executives in the event that they lost 10% of their current budget, what would they cut.  Nearly six out of 10 executives said they would cut consumer education if budgets were reduced.  This points to a serious lack of communication between utility executives and consumers.

If utility executives had to deduct 10 percent from their budget, what would they reduce

• 23% Energy Efficiency / Water Conservation
• 26% Staff / Employees
• 20% Infrastructure
• 21% Upgrading Our Technology
• 33% Consumer/educational programs

Barriers to change

94 percent of utility executives identify government policy and uncertainty about regulations in the future as the most important barrier to infrastructure investment.  Consumers also seem to agree that government or a lack of policy is hindering infrastructure investment.

Which if the following do you believe are barriers to infrastructure investment ?

• 54% Government regulation that delays investment
• 45% Lack of clarity of government regulation
• 40% Uncertainties about the future of the industry
• 34% Difficulties with prioritizing where to invest
• 32% Lack of public funding
• 25% Lack of private funding
• 19% Need more time

Role of technology

Utility executives were also asked in the event that their budget was increased by 10%, where would they spend the money,  More than half of utility executives said that the very first thing they would do with more budget is spend it on technology.  Of these 68 % said they would invest in IT services and 58 % would invest in the sensor web often termed the “Internet of Things."

If Utility executives had an extra 10 percent in their budget, what would they spend it on ?

• 51% Upgrading Our Technology
• 43% Infrastructure
• 41% Staff / Employees
• 39% Energy Efficiency / Water Conservation
• 25% Consumer / Educational Programs

Big data and analytics

A large proportion (74 %) of utility executives feel big data analytics is going to be central to infrastructure modernization efforts. Six out of 10 consumers also believe big data drives utility operational efficiency.

According to this survey 46 percent of utility executive respondents feel prepared for big data analytics today. ( This is surprising.  A similar recent Oracle survey found that utilities are more prepared to manage the huge volumes of data that smart grids are capable of generating in 2013 than they were one year ago, but the majority still say they are unprepared.  In 2013 17 % of respondents said they are completely prepared compared to 9 % in 2012.) Consumers are very concerned about big data readiness, with only two out of 10 responding that utilities are prepared to take advantage of big data analytics today.

Utility executives liekly to invest in technology would choose

• 68% IT services (programs, software, hardware, etc.)
• 58% Internet of things (i.e. embedding sensors in physical objects)
• 45% Sensors
• 38% Meters

In summary

The Itron study found that there was near unanimity among utility executive and consumer respondents that the industry needs to transform itself. It found that consumers want help from utilities to better manage their resources and that policy and regulation are perceived as critical, but they are seen by both utility executives and consumers as a barrier to progress.  Communications between utilities and consumers needs to improve. Most of respondents see technology as the key to the transformation the utility industry needs to go through.

Sensor web: drifting buoys

I blogged earlier about the sensor web that is taking shape.  It involves satellites and ground sensors.  I blogged about very inexpensive nanosatellites, typicaly 10 cm x 10 cm x 10 cm,  that are dramatically reducing the cost of sateliite imagery.

Drifting buoys are less than 22 feet long, weigh no more than 100 pounds, and typically survive on the open seas for an average of 450 days. Each drifter consists of a surface buoy and an underwater drogue attached by a tether. The thermometer, pressure sensors, batteries and other electronics reside in the buoy.  There are about 1,000 drifting buoys on the world's oceans. The modern drifter weighs about 44 pounds with 16-foot drogues and cost around $1,700 each.

Roughly every hour, the drifter uploads data to a satellite passing overhead which in turn relays the observations back to an Earth receiving station.   Drifters measure sea-surface temperature and ocean current data as well as their location. For example, their observations help in tracking the impact of El Niño and La Niña on global ocean currents. In addition NOAA will sometimes deploy drifters in front of hurricanes to improve forecasts and investigate how the ocean and atmosphere interact within and around the storm.

The 1,000 or so drifters in the oceans collect about 30,000 observations per day, which are processed by NOAA's Drifter Data Assembly Center in Miami and archived for research and placed on the Global Telecommunications System, where the National Weather Service and other forecasters can use them for weather prediction.

The map shows the location of all drifters as of April 26, 2013. (Credit: NOAA Climate.gov, based on drifter data from the Atlantic Oceanographic & Meteorological Laboratory.)

"3D" interoperability for the smart grid

At the inaugural meeting of the SGIP John McDonald, Director of Technical Strategy and Policy Development at GE Energy and Chair of Smart Grid Interoperability Panel (SGIP) made a strong case for why interoperability is even more important now than it was in 2009 when NIST initiated SGIP.  Back in those days vendors like GE were selling products and interoperability was to a large extent the customer's problem.   Since then there has been a "solution revolution" at GE where GE now focuses on  understanding the customer's business problem and then builds a solution by integrating components from across GE's many divisions and product groups to solve the problem.  But after doing this, they quickly realized just how critical all these components working together was to delivering successful solutions.  They also recognized that product maturity was not what they expected.  The result was that at GE interoperability became even more important than it had been back in the days when they sold products, which is one of the reasons that GE is a major supporter of SGIP and a Conference Sponsor.

In an interesting EnergyCentral webcast yesterday, Pat Corrigan Vice President, Public Utilities at Elster, Richard Gray, Manager of Enterprise Architecture at Kansas City Power & Light and Doug Lambert, Multispeak Program Manager at NRECA gave their perspectives on interoperability for the smart grid.

Pat Corrigan provided an overview of some of the challenges facing the utility industry as it moves toward implementing smart grid technologies.  Perhaps the most important challenge is simplifying the complex web of devices, communications protocols, and applications such as SCADA, outage management (OMS), billing, DSM, AMI, and other applications.  In the past each application represented a silo and there was very little interoperability between application silos.  This is the world that Pat calls 2D.  Each application uses its own protocols to talk to its own devices.  Whatever integration there is between application silos is point to point and often involves custom programming that has to be retested whenever a new version of either of the applications is released.

Pat outlined a vision of the future, what he refers to as "3D", where applications, communications protocols, and devices all interoperate using standards and multi-device and multi-communications support built into devices and applications.  For example, Elster has developed an integrated smart meter and radio called REXUniversal that supports a number of different radio protocols (multi-radio) as well as full support for standards such as IEEE 802.15.4g and Zigbee.

I hope this is indicative of a trend because if all vendors take the approach of building devices that support standards such as those from the IEEE, IEC CIM, Multispeak, OGC, and those in the SGIP Catalog of Standards as well as multiple communications protocols, developing smart grid solutions will become easier. 

Standards enable devices from different vendors to be used interchangeably, thus encouraging competition and ultimately reducing prices.  The internet is a classic model because it is a suite of standards from several standards organizations like the IETF, IEC, and W3C and is not "owned" by any vendor.  But vendors can develop devices and applications that interoperate with other devices and applications from other vendors because they all use TCP/IP, HTTP, HTML, JSON and other internet standards.

802.15.4g#sthash.dmvFDCFk.dpufZigbee and can interoperate with any wireless network commonly used by utilities for connecting smart meters to aggregators and my consumers to connect a home network.

Automating image-processing work flows for the Earth observation sensor web

Just about every industry from construction to utilities to roofers to insurance is being impacted, and often transformed, by the availability of high resolution Earth observation (EO) imagery from satellites, planes and helicopters, and UAVs.  Initially 2D, then 3D and in 2014 4D (2D/3D + temporal) imagery is available from many sources.  Imagery from international satellites,  low cost satellite constellations and UAVs is poised to dramatically reduce the cost of EO imagery.

An exponentially increasing volume of EO imagery is being captured every day from a variety of devices. The Committee on Earth Observation Satellites (CEOS) reports 286 Earth observation devices in orbit. The next Digital Globe satellite Worldview-3 scheduled for some time in 2014 will be capable of 31 cm resolution and can store two terabits of data between downloads. Two start-up satellite companies have already started putting satellite constellations in space, that promise to provide high frequency revisits to every point of the Earth's surface at low cost.  Planet Labs and Skybox Imaging each plan to launch constellations of 24+ low cost satellites that will capture high-resolution imagery of every spot on earth many times per day.  They promise to deliver the first ever HD-video of any spot on earth,  allowing users to track changes—from traffic jams to deforestation—in near real time. 

A few years ago I remember that the total amount of imagery being downloaded daily from all EO satellites was estimated to amount to about a terabyte.  Now a single satellite could be responsible for a terabyte daily.   If you add advances in aerial photogrammetry and the rapid development of UAV-based photogrammetry, we are looking at huge volumes of imagery every day, probably approaching a petabyte.

All of this data requires processing before it is made available to customers and end users.  Currently this processing is time consuming and a significant proportion of it stil involves manual and semi-automated processing.  This results in delays between the time the data is captured and when it is made available to customers which can range from days to weeks or even months.  Customers are becoming impatient with the time it takes to process data and are increasingly pushing for near real-time availability.

At the PCI Geomatics Reseller Meeting in Ottawa Wolfgang Lück, whose company Forest Sense cc customizes image processing workflows and provides remote sensing training, gave an overview of the sensor web and how automating image processing is enabling near real-time imagery availability.

Wolfgang says that most of the sensor web including different types of satellites, ground-bases sensors, mobile devices and satellite monitoring systems called sentinels are in place.  

• EDRS Geostationary Relay and supercomputer
• Wide area sytematic Optic/SAR monitoring
• VHR SAR
• Optical VHR/hyperspectral
• Field sensors
• Mobile devices
• DRS (earth receiving station)

According to Wolfgang, the communications protocols and the standards (for example, the OGC Sensor Observation Service) are in place for these systems to talk to each other.  Most of the satellite systems are in space, with just the sentinel systems still needing to go up. These are wide area systematic Optic/SAR monitoring systems being launched into space by ESA.  On the ground there are receiving stations (DRS), field sensors, and mobile devices carried by people.  The mobile devices capture as well as receive information.

Imagery requires involves complex processing to make it usable.  With the huge volumes of data that the sensor web is capable of capturing this processing has to be automated if the data is to be available in near real-time at low cost.  There are many actors that need to be accounted for, but the most important actors are end user customers (machines or human beings) because customers determine the products that are to be delivered.

Typical image processing workflow

There are different work flows depending on the sensors that are used and the final product requested by the customer.  To provide a feel for just how complex this processing is and the challenges involved in automating these work flows,  Wolfgang went through a typical work flow beginning with raw satellite imagery and following through to producing an orthorectified image suitable for classification, for example to identify different types of vegetation.  Even for this pretty basic image processing work flow, there are many steps.

1. Ingestion of data from pick-up point
2. Binary data to supported format conversion
3. Relative radiometric correction and artifact removal
4. Band alignment
5. DN to ToA reflectance conversion
6. Haze cloud and water classification
7. Auto-ground control point collection
8. Orthorectification
9. DSM generation
10. Topographic correction
11. Spectral preclassification
12. Level 4 product generation
13. Image compositing / mosaicking
14. Delivery of data to pick-up point

In the past many of these steps were manual or semi-automated.  The important breakthrough that Wolfgang was able to demonstrate is that he has been able to automate the entire process, which means that end user products, whether orthorectified images, digital surface models or classified vegetation maps are available much faster and at lower cost.

From Wolfgang's perspective the key to enabling this to happen is an image processing platform that provides a modular architecture, robust scripting, "big data" management, support for a broad range of sensor models (cameras),  support for Linux as well as Windows, comprehensive image processing functionality, the ability to incorporate custom leading edge algorithms -  integrated with an accessible user interface and API.  In addition because this level of automation pushes the envelope, responsive support from the vendor is also essential.

In the example Wolfgang used a scripting language, Python or EASI, to implement a work flow primarily comprised of PCI-Geomatica modules, but also custom components implementing leading edge algorithms from current research.

The first step is format conversion.  The data often comes in some proprietary format which the processing engine may not support.  A converter is required to bring it into a format that the processing engine supports. 

The next step is radiometric corrections and artifact removal.  Forest Sense supports a wide range of systems put together by emerging space nations, such as African or Middle Eastern countries.  In the example, you can see two lines lines are artifacts.  New gains and biases have to be calculated for those lines and applied to the image.  Other examples are random noise detection and dropped line removal.

Another problem is systematic noise.  A fast Fourier transform (FFT) reveals low a certain pattern in the high frequency domain which is characteristic of  systematic noise.  We can detect that automatically, do a FFT, create a mask to filter it out, and then do a reverse FFT to reconstruct the corrected image.

The next correction is the flat field correction that ensures that the sensors are calibrated correctly relative to one another.

The next step is band alignment.  The different bands need to be aligned because they are acquired at slightly different angles by different sensors.

The next step is total atmosphere reflectance correction.  What the sensor actually gets is relative voltage, and that is converted to relative radiance, which need to be normalized to total atmosphere reflectance.  This allows images acquired at different times, or from different sensors to be compared and to be used for quantitative analysis.   But the effect of the atmosphere is still there.  We can do atmospheric correction, but it is a bit of a black art and applied incorrectly can introduce artifacts.  For this reason, many people in the domain prefer to use total atmospheric reflectance data because they can trust it. 

At this point the image can be classified.

The first step in classifying the image is haze, water and cloud classification.  To do this requires both blue and red bands. These bands correlate with one another very highly, but the blue band is affected stronger by haze and water vapour.  What is called a clear sky vector can be applied to the image to calculate the haze vector and remove haze.  The right of the side shows the haze that is removed from the image.

The next step is collection of ground control points (GCPs) so that accurate geolocations can be calculated.  Unlike many systems which collect GCPs even over water or cloud which introduces errors, Wolfgang only collects ground control points over land. 

GCPs enable automated orthorectification as well as extracting digital surface models (DSMs) automatically from stereo pairs. 

The next step is preclassification. Spectral preclassification requires topographic normalization using a digital terrain model (DTM).  The image has to be preclassfied into different surface types that have different BRDF (bidirectional reflectance distribution function) properties.  Light reflects in different directions at different intensities on different surfaces at different wavelengths.  This has to be done using bands that are not affected by topography, which makes this step quite complicated. 

The slide shows a Landsat image of a hilly terrain with a lot of shade and exposed slopes which need to be normalized topographically.  Traditionally people have used a range of vegetation indices, such as the normalized difference vegetation index (NDVI).  From comparison with the DTM, it is possible to see that the NDVI is affected by topography.  Although it is a ratio between bands, it is still possible to see shady and sunny slopes.  The reason for this is that there is a lot of scattering in the blue band and shorter wavelengths but no scattering in the red and longer wavelengths so the ratio between the red band and the infrared band is not a pure ratio because the red band is affected by scattering. 

The TVI index was developed to be  resistant to topography.  Of course it doesn't work in areas of full shade.  A DSM is used to to calculate which areas are in complete shade so they can be masked out.  At this point all of these techniques can be applied to generate an automatic spectral preclassification of the image which then can be used for  topographic normalization.  The skylight adaptive topographic normalization is based on techniques that Wolfgang developed using all of these preprocessing steps.  In the corrected image it is still possible to still see some shady areas, but most of the shadow has been removed. 

The next step is image compositing and mosaicing and the generation of Level 4 (geoinformation)  products.  These products might include vegetation indices, biophysical parameters, masks, pixel-based classifications or object-based classifications. 

The final step is to package the data and deliver the data to the customer, typically by transmitting the imagery to a pickup point where it is catalogued.

At this point the product is available for the customer to collect from the pickup point.

Wolfgang demonstrated that downloading raw data and transforming it into something usable by a farmer, construction contractor, or insurance adjustor requires many steps and a lot of processing.  This example of a typical image processing work flow shows just what nearly a petabyte of the raw imagery captured daily must go through to create something usable by the end user customer (people or machines). To get this to customers in near real-time is incredibly computationally intensive and involves rapidly processing massive volumes of data.  This is why algorithmic performance is critical and why multiple processors with hyperthreading and most recently GPUs (graphics processing units) are being harnessed to improve throughput.