Shopping for a Digital Photogrammetric System
"The factory of the future will have only two employees, a man and a dog. The man will be there to feed the dog. And the dog will be there to keep the man from touching the equipment" - cited by Prahbu Guptara.
I hope that automation in photogrammetry will not result in comparable consequences. I recently traversed far and wide to look at Digital Photogrammetric Systems (DPS's) and was highly impressed with what I saw. Systems showed remarkable ergonomic quality, and there was a high level of professionalism among the personnel I met. It is a miracle that I managed to avoid investing in a few of these systems for personal use. But even with excellent software Photogrammetry can be a complex task when things go wrong, requiring experienced handling from a user. It could never be imprisoned by automation, and so I do not see this utopia of a photogrammetric lab populated by dogs and men
To Whom We Owe All These
When I started my carrier in Photogrammetry several years ago, the heroes were Von Gruber, Kasper, Schermerhorn, Helava, Ackermann, etc., etc. As digital photogrammetry is further empowered through automation our "heroes" should include some icons of the IT industry. Of special mention is Intel founder, Gordon Moore, who ordained that the number of transistors on a microchip doubles every 18 months. Such multiplication of silicon power has allowed computers to improve with steadily falling prices, a major inspiration in developing totally digital photogrammetric systems. Today we have the 450 MHz Pentium (the 550 MHz was launched February 17, 1999) which when installed in dual format makes automated triangulation look like child's play.
Automation and Alliances
Automation is being employed increasingly, notably in Aerial Triangulation where it is now in everyday use. Automated feature extraction has stalled and will probably remain in abeyance. Prices quoted recently show a remarkable downward trend, and sales of licences have been boosted by market expansion into new areas. Protective instincts remain as evidenced by extensive and, sometimes, confusing strategic alliances. In many cases DPS systems have tools licensed and embedded from rival systems. Some consolidation is in place with the announcement of a new venture between Intergragh and Zeiss. Similarly, but with cosmetic consequences, the parent company of GDE Systems has changed again with the take-over of GEC Marconi by British Aerospace.
No Panacea for Selecting a System
I have discovered there is no panacea for selecting a Digital Photogrammetric System. The way you go about this will depend on your circumstances, e.g.
 | what do you have already |
| what kind of job do you want to do |
| how much money are you prepared to spend |
| what manpower flexibility do you have |
Are Digital Systems in Transition?
Not really! But people are still asking for more facilities and easier predictability of results, no matter the system employed. Many insist that, because of the word "digital", DPS promised automation but this has been elusive for certain production lines in spite of substantial investment. Focus has also increased on the processing of directly acquired, very high spatial resolution digital images. A few air-borne camera projects hold this promise but it will be a couple of years before we have the first serious commercial systems. The image-processing engine of a DPS is vital for image presentation, visual analysis, strain-free stereo viewing, and optimal manual measurement. The quality of this ‘engine’ may have separated systems in the past but not any more.
UNIX vs WINDOWS
The Sparc ultra leads the way, but the 450MHz Pentium is not slow one bit, especially in dual mode. In the end it depends on what you need to do your job. Speed has been an obsession in digital photogrammetry for a long time; in fact the advent of digital photogrammetric systems relied heavily on the availabilty of fast CPUs at affordable prices. Real-time image panning and rapid generation of DEMs are very demanding tasks and the available technology has always been a major issue in the design and implementation of all commercial digital photogrammetry software. This was what separated UNIX-workstations and PCs five years ago. Nobody worries about this anymore.
User Friendliness
It may be easier to manage a PC than a UNIX system, but this applies to new comers and infrequent users. Operators seem to master UNIX applications easily and would not bother about it anymore. Again, no matter how user-unfriendly software may not be, experience shows again that operators acquire the expertise needed fast and may actually be comfortable with the knowledge that emergency-operators cannot easily take over. A typical University-teacher, who comes into the lab once a month for tutorials, would prefer the software to be rather obvious. However, the most important issue is that software tools be robust, effective, fast, and bug-free.
Brand Loyalty
It may be falling a bit but it is still there. There are many conservatives in this business that are tied religiously to one vendor. They will make the odd purchase here and there, or invite you for a benchmark to ‘justify’ their conclusion, but their minds are made up!