Since its inception how much has photogrammetry gained relevance as a tool to generate and analyse geospatial data?
Ian Dowman, the President of ISPRS shares his thoughts on the developments in the realm of Photogrammetry, along with key achievements and future plans of the ISPRS in an exclusive interview with GIS Development.
Since its inception how much has photogrammetry gained relevance as a tool to generate and analyse geospatial data?
Photogrammetry has developed slowly since it was first used in the second half of the nineteenth century. Its initial use was for mapping and for the recording and measurement of buildings. These applications continued as the primary applications until the introduction of computers and analytical plotting instruments in the 1970’s. Since then the value of geometric models has been realized in a wide range of other applications from computer animation, construction of virtual 3D models, analysis of medical imagery to airport security and disaster management. In all of these applications the geometry has become important: to obtain reliable measurements to create realistic models and to enable different types of data to be merged and used together. The emergence of geographic information systems and the world wide web has encouraged this trend, providing an enormous demand for data.
Can you highlight some recent research and development, which have added value to the field of photogrammetry? What are the future trends in technological development in this realm?
Photogrammetry has always been technology led. The mapping industry has responded and embraced new technology for its own purposes. Undoubtedly the biggest development in recent years has been the introduction of digital photogrammetry; the most important single result of that has been the ability to generate digital elevation models (DEMs) automatically. The interest and use of DEMs has been strengthened by the availability of high quality stereoscopic data from space, and the emergence of techniques for DEM generation from interferometric SAR. It has been possible to cover large areas with quite reliable DEMs, (although the methods are not yet perfect); the Shuttle Radar Topography Mission (SRTM), Nextmap Great Britain and the generation of global DEMs from SPOT 5, being outstanding examples. The use of LiDAR has further extended the range of DEMs, which can be generated from airborne platforms. The use of Unmanned Autonomous Vehicles (AUVs) may also have an impact on the wider use of photogrammetry.
Research into automatic extraction of features, particularly buildings, has been going on for many years, but a reliable, robust solution has not yet been found. Developments in technology have greatly improved what can be done with photogrammetry; for example 3 line digital cameras provide significant redundancy in geometric solutions and also provide 3 views of every point on the ground, which makes true orthoimages production possible. The fusion of different types of data, such as LiDAR, can also help the process. A combination of greater computer power and automatic point detection has led to very powerful software for automatic aerial triangulation, which is particularly useful for close range photogrammetry when self-calibrating bundle adjustment is required. This can be provided on a laptop and is therefore highly portable and near real time.
Real time photogrammetry from airborne platforms is also close and this has important applications to disaster management, for example. A further technological development, which has importance in the development of photogrammetry is the internet. The capability to deliver data and some form of processing over the internet will extend the use of accurate 3D data, and create new applications.
How has the relation between photogrammetry and GIS industry evolved over the years?
GIS has helped the development of photogrammetry because the use of GIS creates a need for data to populate the database. One important data type has been imagery, which of course needs to be orthorectified and up to date. Developments such as Google Earth require an enormous amount of data to provide global coverage and to provide 3D models for urban centres, as is the case with some cities at the moment. The importance of reliable, contemporary data cannot be emphasized enough.
What are the key hurdles faced in the wide application of photogrammetry?
The hurdles can be divided into two types: technical and economic. The improvement of automatic techniques is undoubtedly a major requirement. The creation of 3D data for the databases mentioned in the previous section cannot be economic unless it can be done automatically, so more research is required in this area. Maybe some lateral thinking is required to find completely new techniques. Automatic registration of images of different types is another essential requirement.
The other type of hurdle relates to the market. Data is still costly and manual processing is also costly. There is a poor understanding amongst many users of the value of accurate 3D data. Filmmakers have recognized the usefulness of photogrammetry in animation, films such as Spiderman are an example, but the photogrammetric community has not generally been involved in this type of work. There is a growing interest in the use of photogrammetry for mapping of cultural heritage, the work done by Gruen in Zurich, and the generation of 3D animations, comes to mind, but often cost is a barrier to the extension of this type of work.
The establishment of interoperable Spatial Data Infrastructures will help with the distribution of data, and this may help the establishment of a more liberal data pricing policy in some areas, which will also help the promotion of photogrammetry.
As more and more high resolution satellite images are coming up in the market at reduced prices, what future do you foresee for the photogrammetry field viz-viz High Resolution Images?
It is unlikely that high-resolution satellite images will replace airborne imagery in the near future because they cannot reach the cm level pixel size, which is possible with airborne platforms. However photogrammetry is still needed to ensure that accurate 3D data is produced from any satellite imagery. There are some key application areas, which will be addressed by new data, particularly the updating of databases, disaster management and agricultural monitoring: all of which require a very rapid turnaround thus returning to the need for better automation.
A very important development in the past 2 years has been the formation of the Group on Earth Observation (GEO) and the plan for a Global Earth Observing System of Systems, (GEOSS). GEOSS will establish an architecture for the delivery of Earth Observation data to end users, concentrating on areas for the benefit of Society. ISPRS will play a role in this, by ensuring that processes are in place to ensure that good quality DEMs and orthorectified imagery is delivered and by assisting in training and outreach activities.
Please tell us about some key achievements of ISPRS in recent times.
ISPRS has a very strong science programme, directed by the eight Technical Commissions. An excellent team of Working Group chairs and co-chairs has been appointed and over 30 workshops have already been held or are planned for 2005. The planning for the eight Technical Commission Symposia in 2006 is well advanced. The Technical Commissions and their Symposia are listed in
Table 1. ISPRS Technical Commission, Presidents and Symposia
Image Data Acquisition –
Sensors and Platforms
3-6 July, 2006
Theory and Concepts of
Handling and Information
University of Vienna, Austria
Societies” 12-14 July, 2006
Vision and Image Analysis
University of Bonn, Germany
20-22 Sep, 2006
Indian Space research
for Sustainable Development”
27-30 Sep 2006
Analysis and Applications
Dresden University of
& Vision Metrology”
25-27 Sep 2006
Education and Outreach
Tokai University, Japan
Symposium 27-29 June 2006
Modelling and Analysis of
Remotely Sensed Data
John L. van Genderen
From Pixels to
8-12 May 2006
Remote Sensing Applications
University of Haifa, Israel
3-7 Sep, 2006
The ISPRS Congress will be held in 2008 in Beijing and that will be the culmination of four years of hard work by many people. Some notable activities of the working groups, which have been already completed are the First International Symposium on Geoinformation for Disaster Management held in Delft on 21-23 March 2005. WG meetings in recent months have been held on LiDAR, Space Technologies SRTM, 3D Virtual Reconstruction and Visualisation of Complex Architectures, Optical 3-D Measurement Techniques and Service & Application of Spatial Data Infrastructure. In the future many more meetings will be held, for example a Workshop on Topographic Mapping from Space in Ankara, Turkey.
ISPRS has also collaborated with IEEE and AARSE to organise a workshop on GEOSS in Africa. Other activities in this area include the preparation of a video to promote photogrammetry and remote sensing, and the organisation of a Summer School for Students.
The ISPRS Book Series was started in 2003 and contains collections of high quality scientific papers, usually presented at ISPRS events, after peer-review. Three Books have been published so far:
- Advances in Spatial Analysis and Decision Making, edited by Zhilin Li, Qiming Zhou and Wolfgang Kainz;
- Post-launch Calibration of Satellite Sensors, edited by Stan Morain and Amy Budge;
- Next Generation Geospatial Information, edited by Peggy Agouris and Arie Croituru.
The ISPRS Foundation was launched in Istanbul in order to attract donations from other sources, and already we have attracted donations of $US50,000. Council has also decided to transfer $200,000 from our reserves to the Foundation as a loan to allow support for worthy activities to start as soon as possible. We still need more donations, and Council and Foundation Trustees are working to make this happen. We are also looking for support from other sources to fund the activities set out above and any suggestions on who may provide such funding will be welcome.
What are the future plans of ISPRS?
In 2004 ISPRS Council set four objectives for the next four years, these are:
- Sustain and develop the scientific programme based on international excellence in research and in collaboration with other international scientific unions;
- Expand the international role of ISPRS by building on our existing links and developing a presence in developing countries;
- Continue the role of ISPRS in education and technology transfer in collaboration with international partners;
- Develop the Foundation and attract $500 000 of funds by 2008.
We have already set in motion activities to realize these objectives and some of these are reported here.
We are participating in the International Polar Year and working with the GeoUnions group within ICSU (The International Council for Science), which is developing projects on Cities and Megacities, Desertification, Groundwater, Hazards and Health. We are also represented on the UN ad hoc study group on the possibility of creating a disaster management international space coordination entity.
We will continue to represent the interests of the photogrammetry and remote sensing communities in international foara such as the United Nations and ISCU and we will be holding tutorials and summer schools at various locations during the next 3 years.
There is an important role for the science and technologies of photogrammetry and remote sensing in many areas of application, and ISPRS intends to ensure that these are recognised and promoted as much as possible.