Dr.-Ing. Juergen Dold
Business Area Imaging and Scanning
Leica Geosystems AG
Q.How do you see your market in the areas of LiDAR and airborne imaging?
images/magazines/2008/sept are getting more affordable. They are now used in areas where they were not used earlier because of cost. For example, oil exploration companies are using orthophotos for planning purposes where they would have otherwise sent survey teams.
Ground survey would have taken months of work to get less information than from airborne data acquisition. I do not think LiDAR is replacing photogrammetry. These technologies are unique in their own way. For example, if you want a fast DTM, LiDAR is the most cost-effective solution and yields high resolution products. LiDAR has opened new applications in areas of utilities, road mapping, environmental impact, biomass determination, etc. Imaging and LiDAR both have their respective applications, and they are even beginning to merge. In city modelling, you can use a terrain model from LiDAR over which you drape an orthophoto. LiDAR is a multi-dimensional driver of applications.
The need for higher resolution images/magazines/2008/sept than those traditionally acquired from satellites is fuelling growth for airborne digital imagery. The race between Google and Microsoft to get imagery in the hands of consumers has also raised awareness about geospatial technology. At the same time, Leica is seeing even faster growth in our LiDAR business as many of our customers have largely switched to LiDAR to derive terrain models.
How do you view the trends in data fusion and packaged datasets?
We create sensors to do both together. Our airborne LiDAR data is integrated with our medium format camera images/magazines/2008/sept, all perfectly georeferenced by the same inertial plus GNSS georeference system. No longer do you have to have a two-hole aircraft and two sets of sensor controls, as needed with a large format camera (LFC) plus LiDAR. With one flight, you get georeferenced image and LiDAR datasets that can then be fused by software.
Leica also sees great progress in fusing terrestrial scanner data with airborne LiDAR point clouds. We have also fused data from thermal sensors, hyperspectral sensors and RGB (colour) imagery. Leica believes that leading edge sensor hardware technology must be combined with equivalent software/workflow expertise to create a total sensor solution. Our BA I&S software groups, combined with the talents of ERDAS, are integral to our industry leading position in airborne sensor technology. Our customers thereby have the flexibility to push the envelope in traditional data deliverables, data fusion and other packaged datasets.
Are photogrammetrists accepting the new digital format of data?
Yes and no. Working on digital data requires changes in a company’s data production workflow. Larger companies producing large mapping projects have already implemented these more capable workflows. For companies that were using film cameras, even those with film scanners and digital photogrammetry capabilities, the heavily modified processing steps of digital sensor imagery becomes a challenge.
For digital sensors, workflow virtually starts on the airplane. With the entire industry now shifting to digital sensors, use of film cameras will decrease dramatically and end within the next few years. Photogrammetrists are keeping up with the differences in scanned digital data from film Vs direct digital data from digital sensors.
The larger challenge to many photogrammetrists has been the new photogrammetry software that makes it easier for someone without a photogrammetry background to competently process imagery.
Q. Was the transition from analogue to digital painful?
Leica started 50 years ago with our RC series of filmbased cameras. In these 50 years, there have been progress ‘blocks’ of 4-5 years each including developments in automatic film transport, lenses, etc. leading finally to the RC30 – the gold standard of airborne film cameras. Leica has always started early. In some cases, we can say we started too early and could have waited longer for industry to be ready for new technology.
We knew that the most innovative sensor solution would be required to fulfill the automation level required by the mapping industry. Therefore, Leica chose the ‘pushbroom’ line sensor concept for its large format airborne imaging sensor. With this sensor, we provide the highest quality imagery for a changing world. Leica introduced the first digital camera ten years ago. Leica has a perfect base for starting new technologies. We have optical, mechanical, software and electronic capabilities plus associations with leading academic and scientific research centres to help us manage the challenges.
You can therefore see that there has already been a ten-year overlap of analogue and digital imaging technology with a few years of film technology left. The transition has been straightforward for Leica as a manufacturer and many of our larger, innovative customers. It has been more stressful for customers with more limited financial and technology resources that help support such a major transition. There has always been changing technology and changing performance. The demand for imagery has gone up. Data providers want to fly more and want multiple sensors on each flight. And, so has the pace of sensor advancements changed. Just two years ago, we released a LiDAR system with a 100,000 kHz pulse rate and today we have the ALS60 LiDAR that delivers 200,000 kHz. Two years ago we released the second generation ADS large format digital sensor. This year, we released the third generation ADS80 that collects images/magazines/2008/sept even faster and uses the newest flash memory technology. Technology is changing at blinding speed and product cycles are shorter forcing our customers to adapt at a similar rate.
What do you do to sensitise your customers?
This is done largely through direct contact with customers. Our people attend conferences and other similar fora to understand our customers and identify the challenges. We do projects with our customers and get to know application and performance requirements trends. The customer education and sensitisation process is very important to us. Leica must always balance the competitiveness of our product lines with the concerns of our customers that their investments sometime seem to become outmoded too fast! Leica has to push the technology forward at a pace the market can adopt. We want the market to push strongly and when it does, our R&D efforts must be ready with the appropriate, reliable technology. It is not just about providing a sensor, but also about providing a solution that the market needs and about providing an affordable upgrade path to keep all of our customers at the leading edge.
Take the example of our ADS with its pushbroom workflow for faster processing of raw data to deliverable images/magazines/2008/sept. Our long term vision is to process acquired data in real time during the flight. Sensors must be smart and processing must be fast and efficient. That is a challenge to the industry, and Leica is ready. We call it “Imagery for a changing world”.
Q. Are we going to go back to photographic frame?
Today, there are solutions that claim they are (single) frame, but in fact they use multiple frames merged into one virtual frame. The pushbroom sensor was designed for satellites because it is the perfect solution for efficiently collecting large area, perfectly co-registered, multi-spectral data. The different line sensor workflow compared to frame workflow requires straightforward training to understand the differences and to benefit from its superior performance. Ironically, many people feel the workflow from film to digital ‘frame’ sensors is almost identical, but this is not the case. The processes that attempt to correlate the multiple images/magazines/2008/sept from the multiple camera heads that shift around from thermal and mechanical instabilities can be harder to learn and produce results inferior to line sensor workflow.
Our customers find that ADS line sensor workflow is much easier and more efficient to work with. Frame sensors require up to 90% forward overlap and up to 50% sidelap and all the data from the multiple heads has to be correlated and mosaicked. A line sensor requires zero forward overlap since it has built-in forward, nadir and rearward arrays that scan a continuous, infinitely long strip. There are also other advantages to central (frame) Vs parallel (line) perspective acquired data. It is about mosaicking and management of data. If sensors are not designed to minimise the processing software workload, the goal of moving workflow onto the sensor in flight cannot be achieved.
Frame sensor technology is still competitive for smaller formats such as 39 Megapixel ‘medium- format’ and below used on smaller projects and with LiDAR. However, as the geospatial industry embraces the benefits of line sensors over frame sensors, line sensors will become the technology of choice for wide area mapping, earth resource and most other large-format applications.