HIGH RESOLUTION IMAGING FROM SPACE

HIGH RESOLUTION IMAGING FROM SPACE

SHARE

A.S. Kiran Kumar
Deputy Director
Sensor Development Area
Space Applications Centre (ISRO)
Ahmedabad

1. Introduction The enormous potential of satellite remote sensing for the benefit of humanity resulted in early adoption of space technology for earth observation activities. Prior to space age (conventionally dated from 1957) humankind did not have a global view of the world in which it lived. The satellite observations have provided us with a more complete, global and sometimes near-instantaneous (most recent) view of Earth system1. It has enabled the rediscovery of our planet through the systematic collection and analysis of enormous amount of data received over the past five decades. Currently, the observation data provided by satellite sensors are regularly used in many fields of application viz. environmental monitoring, navigation, weather forecasting, communication, etc. Observing and understanding the Earth system more completely and comprehensively will expand worldwide capacity and means to achieve sustainable development and will yield advances in many specific areas of socio-economic benefit, including:

  • Reducing loss of life and property from natural and human-induced disasters;
  • Understanding environmental factors affecting human health and well being;
  • Improving management of energy resources;
  • Understanding, assessing, predicting, mitigating, and adapting to climate variability and change;
  • Improving water resource management through better understanding of the water cycle;
  • Improving weather information, forecasting, and warning;
  • Improving the management and protection of terrestrial, coastal, and marine ecosystems;
  • Supporting sustainable agriculture and combating desertification;
  • Understanding, monitoring, and conserving biodiversity

With the availability of a variety of launch vehicles suitable for lifting micro to macro satellites into LEO or GEO orbits, satellite sensors were designed to perform various kinds of earth observations. Depending on the application requirements, sensors with grater spatial, spectral, radiometric and temporal resolutions were developed. Also different kinds of non-conventional imaging techniques like synthetic aperture radar (SAR), synthetic aperture imaging lidar (SAIL), Interferometric Imaging Systems, etc were also developed. The optical and infrared sensors currently used for earth remote sensing applications include sub-meter resolution imagers, hyper-spectral sensors, wide-swath sensors with high temporal resolution.

With these sensors it is possible to observe any given region of the earth with greater spatial, spectral and temporal resolution.

These observations have enabled a wide range of applications covering2.

  • Cartography at every scales in 2D and 3D: The demand for good maps arises from many different needs, e.g., the regional development, the town and country planning with roads and rail tracks, parks and forests, water supply and other installations.
  • Land cover inventories: For agriculture, forest management, buildings, observation via satellites offers unique possibilities.
  • Closed waters, open seas, humid zones surveillance: This area offers certainly one of the biggest challenge of the century, as keeping the waters healthy could well become one of our priorities.
  • Installation of communication equipments: For instance, the installation of new telecommunication and television relays requires extensive study of topography, terrain occupation and population density.
  • Agriculture aids and management: Images of cultures could give information on expected yields (crop evaluation) or for the farmer it can give information on demands for fertilizers, for watering or for chemical treatments.
  • Disaster management, Analysis of changes, etc:

2. Scenario Till Now
NOAA-A/Landsat-1 satellite, which was launched in orbit in 1972, was the first one to target systematic coverage of earth surface from space. It has been the true forerunner with multi-spectral observations in the visible and infrared channels having resolution in the 100-m range. It has paved the way for much better satellite sensors with greater spatial and spectral resolution. SPOT family of five satellites with the latest one launched in April 2002 has helped in developing the commercial market for space imagery of the Earth. IRS family of ten satellites which started with IRS-1A in 1988 with a spatial resolution of 72 meters, improved the spatial resolution to 5 meters with IRS-1C in 1995 (Between 1995 and 1999 till the launch of IKONOS this was the highest spatial resolution data available in civilian domain) and further improved the resolutions to 1 meter with Technology Experiments Satellite (TES) in 2001. This was followed up with first dedicated high-resolution stereo mission in Cartosat-1 with 2.5-meter resolution. The latest addition to this – Cartosat-2 with a spatial resolution of better than one meter – was launched into orbit in January this year. As can be seen, the geometrical resolution has continuously improved, for instance, from 100 m for the first NOAA-A satellite to 10 m for NOAA-K or from 10 m for SPOT 1 launched in 1986 to 2.5 m for the recently launched SPOT 5, or 72 meters from IRS 1A launched in 1988 to 1 meter of Carto-2 launched in January 2007.

Other satellites also have been launched over the past 7 or 8 years with performances down to 0.6m resolution but with less wide field of view than SPOT or Landsat. Some of the missions like IKONOS and QuikBird with imaging resolutions of 1m and 0.6m respectively were entirely private funded. A significant event happened during this period is the development of low cost micro satellites with advanced payloads for variety of remote sensing applications. The low cost of micro satellite system allows the launch of multiple satellites (either identical or different) forming a constellation, which greatly enhance the capabilities of the system. Surrey Satellite Technology Ltd (SSTL) at the University of Surrey (UK) has developed and launched a series of micro satellites with advanced Earth observation payloads. In October 2005, SSTL (China) launched Beijing-1 micro satellite. Beijing-1 will provide the Chinese government and commercial users with information on agriculture, water resources, environment and disaster monitoring throughout China. The satellite will be used extensively for monitoring urban development and pollution, especially in the lead up to the 2008 Beijing Olympics, and to generate digital maps of China using the high-resolution panchromatic imager. The RapidEye satellite system to be launched in near future will have five identical earth observation satellites flying in a formation. The orbital formation of satellites ensures total coverage of the earth in a day.

The observations in visible and near infrared region of the Electromagnetic spectrum is affected by weather and it is not possible to carry out imaging over cloud covered regions. This limitation is overcome by use of Microwave region of Electromagnetic spectrum and this provides all weather observation capability. This however requires Active Sensors, which carries its own source of EM energy for scene illumination. Over the last few decades this technology has developed to provide data with spatial resolutions close to few meters. There have been Satellites launched with Synthetic Aperture Imaging Radars operating in X band capable of providing High resolution data.

SeaSat mission launched in 1978 was the fist SAR mission with a spatial resolution of 25m. SAR payloads of ERS missions (30m resolution), RADARSAT-1 (25-100m), RADARSAT-2 (2-100m), JERS (18m) are some of the significant developments in this field. Cosmo–Skymed (Constellation of Small Satellites for Mediterranean basin observation) mission which has been launched very recently is an Earth observation program of the Italian Space Agency (ASI) developed by Alenia Spazio, a Finmeccanica company, as prime contractor. It is a constellation of four SAR satellites flying in a formation providing frequent coverage with a spatial resolution of 1-100m. The system will monitor the entire globe and the Mediterranean area in particular, providing information for a number of applications thanks to the high resolution of the images acquired, the reduced revisit times over the observed sites and the speed with which the data will be made available to the users.

INDIA is developing a Radar Imaging Satellite (RISAT) capable of providing about 3-50 meter resolution depending on the mode chosen. The satellite is designed for operation in Spotlight, ScanSAR & Stripmap imaging modes. Twenty years ago only three countries, US, USSR and France operated a small number of earth observation satellites and currently over twenty countries operate earth observation sensors on 60 platforms.

Together with the improvement in resolution and size of images, comes the increase in data volume to be transmitted from satellite to the ground stations. Improvement in the data compression algorithm, however limits the growth in data quantity to be transmitted. Also it is worthwhile mentioning that the size of the on-board memory increased over time tremendously and with the progress in memory chips, the volume and power consumption of these on-board memory have greatly decreased.

The Earth stations necessary to receive the data transmitted by the satellites have also decreased a lot in spite of data rate increase, because of better antennas and better transmitters on board satellites and better receivers in the ground stations.

3. Trends In High Resolution Observations From Space
Geometric resolution: With a resolution of about 100 meters in 1972 the geometric resolution of electronic imaging from satellites has shown significant improvements. In 1986 the resolution improved to 10 meters (SPOT-1). In 1995 the resolution improved to 5.6 meters (IRS 1C). In 1999, the resolution improved to 1 meter (IKONOS). It saw a further improvement to 0.61 meters through Digital Globe’s QuickBird. While a number of satellite missions are being planned for resolutions between 0.2 to 0.5 meters nobody has announced a programme for 0.1meter resolution electronic imaging system. It can be seen that the improvement in resolution is taking place at approximately one order once in 13 years. From 100 meters to 10 meters in 14 years, 10 meter to one meter in 13 years. If this trend were to continue one would expect to see a 0.1 meter resolution Imager before 2012. Table 1.0 shows satellites with one meter and better resolution operated/planned by different countries3.

The improvements in resolution demand significant technological advances in the areas of optics, detectors, electronic signal processing, storage, and transmission systems. Data rate and compression: As the spatial resolution increases the data rate becomes very high and in view of the limited bandwidth available in X band, which is currently the most used band, it becomes necessary to adopt compression and other methods of data reduction. As the spatial resolution increases, the imaging strategy also undergoes a large change. As the area of image covered at very high resolutions is very small the satellite is equipped with a high degree of maneuverability so that spot images are taken which lie within a certain field of regard of the satellite. This also implies that continuous imaging is no longer adopted and the time required for maneuvering can be effectively used for reducing the transmitted data rate by adopting compression and deferred transmission. Typically the maneuvering time could be in the range of 10 to 20 seconds while the actual imaging could be 5 to 10 seconds.

In addition the deferred transmission could also provide opportunity to carry out compression of data based on the analysis of the entire frame and enable significant reduction in data rate. With the increase in processing capability available onboard one could consider extraction of specific information instead of the entire data being transmitted. One could consider transmitting only the processed results for near real time transmission while sending the entire collected data at a later point in time for detailed studies.

3.1. Satellite Size

The first observation satellites, NOAA was in the 2 ton range. As the resolutions improved swaths became smaller there was a need for agile satellite to facilitate spot image acquisitions. This agility would effectively enable picking images of required locations within a certain field of regard. Such agility demanded that the satellite size be reduced. IKONOS, QuickBird, EROS, Cartosat-2 are examples. However there is conflict in this approach as one seeks to improve the resolution further. Higher resolution demands larger optics diameter hence the size and weight of payload and thereby the satellite to increase its size substantially making it difficult to achieve higher agility.

More recently, the development of small launchers, sometimes derived from reconverted missiles has allowed or pushed the development of smaller spacecrafts. A number of mini satellites and micro satellites (in the 100–120 kg range) with interesting performances have been launched. Micro satellites have been in existence for 10 or more years providing some images of the Earth but their performances in terms of image quality or image production capacity have been rather limited. The new generation is expected to be far better and must be able to produce images suitable for science and applications.

The concept of using micro satellites for high-resolution imaging could see a significant boost when the cost comes down significantly. It would be possible to put micro satellites to provide very high resolutions of the order of 0.1 to 0.2 meters by operating them at very low altitudes of about 100 kilometers. Short life of very low orbits could be more than offset by lower costs.

3.2. Image Or Information?
The final user is generally not interested in getting an image of the Earth taken by the satellite but in the information it contains. The future should see development of more sophisticated means of extraction of the information from the image, taking into account the needs of various users and making use of data banks allowing comparison between two images for instance to detect changes between two periods.

The increase in the number of LEO satellites would lead to time saving, i.e., to shortening the time to get the image. With the decrease in size, and consequently in cost of satellites, this can be expected to be economical also.

3.3. GEO Platform
Earth observation satellites in geostationary orbit from where they are able to observe/ monitor a large part of the globe continuously. Current technology would permit realization of imaging at about 10-20 meters from Geo platforms. Such a system apart from providing useful data would also enable deciding which portion of ground needs to be monitored at a very high resolution at given point of time. But for all satellites working in optical wavelength, clouds affect the ability to take an image. In the future, any Earth observation satellite will have to work in close link with meteorological satellite data to optimize the image capture.

Active or passive: One could see the adoption of microwave imaging techniques to optical domain for improving the resolutions to less than 0.1 meters. The active imaging techniques would receive a big boost with the availability of low cost micro satellites. A constellation of these satellites could be used to realize different functions demanded by active imaging system.

3.4. Earth Observation Satellites: A Possible Scenario For The Long Term

  • Constellation of mini (and/or micro) satellites in Low to very low Earth Orbit (LEO).
  • Geostationary satellites
  • On-board sophisticated compression techniques
  • Sophisticated tools for processing data

4. Conclusion
High resolution imaging from space has seen a steady growth with an improvement of one order resolution once in 13 years over the last three decades. Advances in electronics, and cost small satellite technology could result in a significant change in earth observation scenario, as one would see satellites operating at very low altitudes for realizing high resolutions. A constellation of satellites working for a single mission could enable active imaging systems providing very high resolutions.

It is expected that a constellation of satellites working as a single system will orbit earth The increased data streams and delivered information will allow decision makers to be virtually present through their remote vision in any region of the world at almost any time. The most advanced countries will have daily access to high resolution all weather data in most places on the globe4.

In addition, possibility of observing the Earth in many more wavelength bands, e.g. infrared, visible light, microwave, different polarizations and in various narrow bands in visible light (super-spectral or hyper-spectral) could see significant growth.

References:

  1. Earth observation history on technology introduction.
  2. EO technological standpoint, Invest-WIKI.
  3. Foreign land imaging satellite programs, presentation to NOAA’s Advisory Committee for Commercial Remote sensing, 1/14/03,
  4. Stoney, Mitretek Systems.
  5. Global Earth Observation trends.