The Whys of Oblique Imagery

An investigation into the nature of perception reveals fascinating details about brain function — and could even earn you some free beer.

By Art Kalinksi

About eight months ago, I was attending a conference in Philadelphia. Right after lunch, I went to my room and turned on the TV to check the news; that’s when I happened across a program that mentioned oblique imagery. Since I work for an oblique imagery vendor, the word “oblique” always catches my attention. The more I listened, the more fascinated I became, because the program was scientifically explaining the reactions that I — and many others — have experienced with oblique imagery. In brief, the imagery is more than just pretty pictures taken at an angle: it really helps perception.

Currently, there are three companies collecting oblique aerial imagery: Pictometry International, Rochester, NY (the firm I work for); MultiVision USA, Orlando, FL; and a recent newcomer, GeoSpan, Minneapolis, MN. While I was the GIS manager for the Atlanta Regional Commission (ARC), we used Pictometry imagery, along with the company’s free software for viewing imagery and GIS data. We also used the imagery with the ArcGIS extension and the ArcIMS integration. During the years that I employed it at the ARC, I observed a response to the oblique imagery overlaid with GIS data that seemed disproportionately positive, relative to the seemingly small improvement over traditional ortho imagery — but I couldn’t explain it.

For more than eight years my ARC colleague, Jim Bohn, and I taught ArcView classes through an ESRI ArcView Learning Center that we set up. Of the hundreds of students that we taught, some used the three-day course as the genesis of a much bigger and more sophisticated GIS, but for some the training lead nowhere. I just couldn’t understand why; it was baffling. I had high hopes that ArcView would take off with police and fire departments, but that traction never occurred.

Then, about four years ago, we were exposed to high-resolution oblique imagery, and everything changed. Police officers and firefighters were captivated by this new imagery and the easy-to-use software. But why? I thought the oblique imagery was neat, but I didn’t yet appreciate that it represented a quantum leap over traditional ortho imagery for most non-GIS users.

One afternoon, Clayton County Commission Chairman Eldrin Bell walked by my ARC office. Eldrin Bell was the former chief of police for the City of Atlanta, and when he retired he was easily elected as chairman. I remember him stopping in front of some sample oblique imagery overlaid with GIS data that I had on the wall outside my office. In a “Eureka!” tone of voice, he said, “That’s exactly what we need in Clayton County, that’s exactly what we need!”

In this ortho view, it’s difficult to identify the structure in the red oval.

In this oblique view, it is apparent that this is a multi-story parking garage, and that the land slopes away from the street.

I went out to talk with him, and the next question he asked was, “What street is that?” pointing to a main road. With that I snapped my fingers and said, “That’s why you need GIS data: without it, this is just a pretty picture.” That day Chairman Bell saw the potential of GIS for his county, and realized that with oblique imagery, GIS could be brought out of the back room into mainstream use.

But what was the magic of oblique imagery? Why was it so compelling to non-GIS users? I came to the conclusion that those of us who work with abstract GIS maps and data are used to looking at the world from a top-down perspective, and ortho imagery follows that model very well. Just as a builder can look at a blueprint and picture the finished building, we are accustomed to looking at an ortho world and can easily make sense of it. This is not the case for the average person. But there was much more to it than that, as I realized in my own “Eureka!” moment.

That Discovery Channel TV program I happened across eight months ago explained the reason why oblique imagery is so compelling. A researcher featured in the program was using MRI scans to measure brain activity and perception. He determined that if one views text that spells a word, the brain activity occurs in the left hemisphere of the brain, and perception is quick and easy. If, however, the text is reversed or upside down, activity occurs first in the right hemisphere, then the left hemisphere.

Click on this image to see what it says…

The researcher noticed that the same effect occurs when one looks at an oblique view of an object. The brain activity occurs in the left hemisphere for oblique views, but in the right hemisphere first for ortho views. His conclusion was that if a familiar object is seen, processing is quick and easy. But if an object is not in a familiar frame of reference or view, perception is slower and more difficult because the right brain is called upon to reorient and make sense of the object.

That finally resolved the mystery and explained why non-GIS people find oblique imagery overlaid with GIS data more compelling than ortho imagery with the same data. To the untrained eye, there was no need for mental reorientation, and the whole process was quicker and easier.

This ortho image shows a glass object…but only the oblique view shows the shape. Click on this image to see what it is.

The other factor was the viewing software. When we were teaching ArcView it took two to three days to learn the basics, and weeks to become proficient. The added problem was that if you didn’t use it regularly, the knowledge evaporated. By comparison, oblique viewing software can be learned in about two hours. Of course, it doesn’t have all the functionality of a GIS application like ArcView or ArcGIS, but it is easy — and it doesn’t evaporate. In fact, the increase in new users was absolutely phenomenal. In Atlanta-area counties, the number of GIS users jumped 10- to 20-fold. I attribute it to two key factors: ease of use and, most important, ease of perception.

You’re probably wondering about the beer part. When I saw the left-brain/right-brain TV segment, I didn’t see the beginning of the show and I couldn’t stay for the end, so I missed the credits. I thought for sure that I could use Google to find the researcher, but so far I’ve been drawing a blank. I called several doctors doing research in similar subject areas, with no leads. I even contacted the folks at the Discovery Channel to get their help, but unfortunately they don’t maintain a database of show content that could help identify the program and source. To complicate the search further, it is possible that this research was just one topic addressed during the show, rather than the subject of the entire program.

I want to invite the left-brain/right-brain researcher to FutureView 08, the October oblique imagery users’ conference in San Antonio. So in desperation, I’m asking for your help. I know that there are many readers with better Internet searching skills than I have; perhaps your social network may lead you to someone who knows what I’m talking about and can help me contact the researcher. The search parameters I have used so far include: Discovery Channel, Learning Channel, spring 2007, oblique imagery, reversed text, mirror text, MRI brain activity, left brain, right brain, learning, perception.

To make it worth your time, I’m offering a prize: a case of Heineken to the first person, over 21 years of age, who can identify the researcher who did this study. If you solve this quest, contact me at akalinski@geospatial-online.com.