Kindly tell us about the origin of the company. How was it formed? What was the vision? And how much of it has been achieved?
CEO, ER Mapper
- Kindly tell us about the origin of the company. How was it formed? What was the vision? And how much of it has been achieved?
I started the company 12 years ago with the fairly simple objective of making high quality imagery technology available worldwide. The invention of the ER Mapper algorithm engine made powerful processing possible on low cost PC and workstation hardware. In order to reduce the cost of product, we needed large numbers of sales, and so we built a reseller network that today is the largest in the world for imagery solutions. We have remained remarkably consistent with our original objectives. Of course, hardware, software and imagery continue to advance. The Internet is very exciting, as it solves the bottleneck of accessing and serving large image mosaics. Looking back, it is interesting to see how the ER Mapper algorithm engine continues to define the state of the art. The ability to provide real time roam and zoom access to large image mosaics has in turn increased demand for these mosaics. ER Mapper is being used to mosaic, pixel by pixel color balance, stitch, and compress thousands of orthophotos into single image mosaics that are getting up in to the terabyte size. I find it remarkable – and very exciting – to see how imagery use has advanced.
- How has been your customer profile changed all through these years? Has it in any way changed your company?
Imagery has become easier to use and more affordable. As a result, GIS and CAD users now represent the largest user base for imagery technology. We see this expanding into wider markets over the next 5 years. Because it is now feasible to serve high-resolution 15cm orthophoto mosaics for entire cities, real-estate companies and other non-technical markets are beginning to use imagery as just another source of information. What we are seeing is a definite trend where the user base for imagery is expanding rapidly. The main way this has changed the company is that we now look at whole solutions, rather than just being focused on image processing. This is why we expanded our product offerings into a whole solution range; ER Mapper (to preparing imagery), ECW (to use imagery inside any application), and the Image Web Server (to serve imagery into those applications using the Internet).
- Your company has placed its bet on ECW. How did this product idea generate? How has it been received?
As the experience of Unisys with GIF showed, you can’t own image formats. We have made a firm commitment in making ECW technology freely available to benefit the spatial community. As I mentioned earlier, ECW is one segment of our complete spatial imagery solutions.
The development of ECW was interesting. We were researching image serving technology (which ultimately became released as the Image Web Server). It became obvious that we had to solve the problem of enabling imagery access from within any application, and serving of imagery into those applications using the Internet. Clearly imagery needed to be seamlessly accessible within image processing products, GIS, CAD, Office applications, web browsers and custom applications. The problem is the sheer size of images. A single high-resolution mosaic covering a city is a TB (terabyte) of imagery. A 1:24,000 scale topographic mosaic covering a state is about 400GB of imagery. In 1998, I was doing some research on wavelet mathematics, which has application for DEM and feature extraction using image correlation. Wavelets also offer advantages for image compression. The outcome of that research was a fundamental mathematical breakthrough enabling Discrete Wavelet Transformations (DWT) and inverse-DWT operations to be performed on very large images very quickly, while only using a tiny amount of RAM. We recently received a United States Patent for the ECW technology.
This provided the solution needed to allow us to offer the Enhanced Compressed Wavelet (ECW) image access and compression technology to the market. Widespread use of compressed imagery means that the image compression, application plug-ins, and Software Development Kits (SDK) have to be widely and freely available. This is why the ECW technology has proved so popular – people can freely download ECW Compressor, ECW Plug-ins, and ECW SDK’s from www.ermapper.com.
As your readers may know, LizardTech claimed that ECW infringed their MrSID technology. The US Federal Court recently ruled that ECW does not in any way infringe on the patent licensed by LizardTech for the MrSID product. We are very pleased to see this matter put to rest. Independent reviews such as consistently give very positive reviews to the ECW technology, and software vendors and users have rapidly adopted ECW technology. As result, we see a very bright future for the ECW standard.
- What is the future of IWS-based application in the emerging knowledge-based society?
Only four years ago, if someone had suggested to me that it would be possible to roam and zoom over TB size images, in real time, using ordinary modem based internet access, I would have flatly said it is technically impossible. Yet today, the Image Web Server does all this, and more.
The logistics involved in providing access to imagery suddenly becomes much simpler. Consider a city department with a 200GB image mosaic, covering a city, that is to be provided to the city’s 5,000 GIS, CAD and environmental engineers. Even compressed, this would be a 20GB image file. Previously, this would entail cutting thousands of CD’s to provide the imagery to individual users – who each would have to have enough disk space locally to store the image. With the Image Web Server, the city can simply store the image on a web server, and serve the imagery out to all the users, directly into their respective applications. As a result, there has been a radical change in the way people use and think about spatial information.
For example, you can zoom into a city, and see the locations of houses for sale, relative to roads, parks and schools. Government organizations suddenly have the means to quickly serve imagery to all their users. Environmental and change monitoring becomes easy to use, with a simple web interface. There are already excellent GIS web server solutions for access into database and vector information. Coupled with the Image Web Server, these provide complete access to all spatial information, regardless of if it is in a tabular, vector or image format.
There is no doubt that Internet based image serving technology is revolutionizing the way we deal with spatial information. It is going to be very exciting to see how the internet oriented knowledge based society changes push this technology in new directions.
- What is ER Mapper’s vision behind product development in association with Map Imagery?
The MapImagery product is a 3rd party imagery application for MapInfo users. It is built using a licensed version of ER Mapper algorithm processing engine, bringing the power and ease of use of our imagery engine to GIS users.
As imagery becomes more affordable, GIS and CAD users are increasingly demanding more powerful imagery capabilities. In the past, the competitive landscape made it difficult for different vendors to work closely together. The amalgamation of spatial information, be it CAD, GIS, database, imagery or whatever, has resulted in much closer working relationships between imagery and GIS companies. We are working closely with GIS vendors and other software developers who are integrating our ER Mapper, ECW and Image Web Server imagery technologies into their products, to provide complete seamless solutions to their customers. This is very positive, as it benefits users of spatial information.
- Asia is said to be heterogeneous in terms of economy, technology, skills and approach. Do you have any specific plans for the region with regard to product or solution for GIS industry? How do you see India as your market?
In my view, the USA is the only truly unified market, where as each of the other markets – Europe, the Americas, the Middle East, and Asia, all have quite distinct country by country differences. Having said that, our headquarters is in Australia, and so we have long enjoyed a successful presence in the Asian markets. The regulatory differences in each country can be quite profound with respect to data availability, and this directly impacts on the growth of the spatial market in each country. For example, the USA makes orthophotos, topo maps and cadastral data available at a very low cost, where as in Australia data access can be quite expensive. In some countries, airphotos are not even available to the general public as aerial images are viewed as secure data. Having said that, Asia is making rapid progress with spatial data availability. Fast Internet access is also becoming more available. I was at a coffee shop in Hong Kong a couple of weeks ago, and was interested to see free 10Mbit optical cable Internet access at the coffee shop.
With data becoming more widely available, the other issue is localization of software. It is costly to translate the 1,000’s manual and software interface pages for “Heavy weight” products such GIS or imagery products. I see a strong growth in “light weight” Internet based applications, with only those functions actually needed by the user. This also simplifies the localization issues, as standard web development tools can be used to quickly develop language specific applications.
India continues to have a preeminent position with spatial imagery due to its high-resolution satellite development and associated infrastructure, and will continue to be very important to our future. It is interesting to see several trends in India:
For example, data availability is tightly controlled in India, with only limited and restricted availability of high-resolution data. This is going to severely limit the India spatial information market until this problem is resolved. Despite this, India of course has very advanced theoretical remote sensing programs as well as the remote sensing centers and NRSA, so India is in a good position in the future. The IT centers in Hyderabad and Bangalore also provide a good basis for the future. In short, I see India will be a world player in the spatial information market – once it resolves the problem of data access (other countries are having to deal with this problem).
- How has the availability of high-resolution satellite imagery affected the Remote Sensing software market in general and your company in particular?
High-resolution satellite imagery has gained a lot of attention, and is vital for some applications. We also see the orthophoto aerial photography market growing very rapidly. This has been driven by the demand for large orthomosaics served over the Internet, and the lowering in production costs for orthophotos. Many users require sub-meter resolution, such as 15cm color orthophoto mosaics for cadastral mapping. Given this booming need, the new fully digital airphoto camera systems being introduced by leading photogrammetry companies are going to drive the price of imagery down even further. One thing is clear – the availability of imagery is increasing, with new applications being developed all the time. We continue to put a lot of R&D into automating the processing of large amounts of imagery. For example, ER Mapper 6 included an automatic color balancing wizard that performs pixel by pixel radiometric color balancing to ensure seamless mosaics.
- How do you see the Remote Sensing industry changing in the next few years?
The various markets are moving into a single industry. Where as before we had airphoto companies, photogrammetry companies, satellite companies, and consultants as very distinct groups, they are all merging into offering whole solutions. The imagery market in turn is merging in closely with the GIS and CAD markets. Office applications were once distinct applications (word processing, spreadsheets and so on), but are now sold as unified office applications. In the same way, the entire spatial information is moving towards unified spatial applications, which will be strongly oriented towards web solutions. On the imagery front, we are starting to see terabyte (TB, or 1,000GB) image mosaics being served on the Internet. It is likely that petabyte (PB, or 1,000TB) image mosaics will be available within 5 to 10 years. Perhaps more importantly, imagery is become just another form of business data, offering unique benefits to all sorts of industries.