#Geolocation: Discover the Digital Sherlock in you

@DFRLab 2017 Global Geolocation Challenge: Kuala Lumpur


On July 30, Maks Czuperski, director of the @DFRLab, challenged the #DigitalSherlocks to use their geolocation skills to identify where he was, based on a few photos taken in the area. Beata Biel (@BeataBiel) took up the challenge and in 15 minutes identified the exact location of Maks in Kuala Lumpur, Malaysia. The following story from Beata reveals how she geolocated Maks using #DigitalSherlocks methods.


It was a hot Sunday early afternoon when I noticed a tweet from Maks Czuperski of @DFRLab challenging Twitter users to guess, based on a series of photos he had taken, where he had been jogging with his friends. I was about to head out for a family dinner but decided to give it a try. And guess what? It took me only about 15 minutes to discover where Maks was. Not just the country or the city, but also the proximate running trail. Being a Digital Sherlock is a lot of fun and you don’t need very advanced tools to become one. The more you practice, the more things you can discover — not just for fun.

So how did I work on the challenge? First, I looked closely at all three photographs, to find any specific features that would give me an idea of Maks’ location. Two of the photos only showed people in a jungle — that was not much of a clue. The third photo, though, depicted the jungle but also a city in the background, with quite a specific skyscraper — higher on one side, shorter on the other. I knew it was a visual clue for which I should search.

Next, I simply went to Maks’ Twitter account, to see the history of his latest posts: maybe they would give me an idea of his whereabouts. And indeed, they did. It turned out that during the previous challenge, he had been posting from Changi Airport in Singapore, which had been geolocated by Julian Röpcke (thanks!). If I hadn’t been that lucky with Twitter, I would have also probably checked other social media profiles of Maks, as well as those of the institutions for which he works.

So I already knew Maks must have been flying from Singapore. I checked the hour when he posted from the airport, then went to the airport’s website and checked what flights were taking off around that time. Most of the airports nowadays show destinations, departure times, and arrival times online — however, you have to remember that the times they show are the local ones, and the time stamps of Twitter posts that you see are connected to your account settings. To check the time in Singapore, I used the World Clock website and then assessed the time stamps of the Twitter posts accordingly.

There were quite a number of destinations Maks could have chosen, so I thought I would try one by one and Google each destination’s name (a country or a city) and the term “skyscraper.” I was lucky: my instincts told me to search for “Malaysia skyscraper” first, and through Google I identified a few Malaysian skyscrapers, the first of these being the Telekom Tower (of Menara Telekom). It looked like the “two-layer” building I saw on Maks’ photo. Of interest: if I had first done a Google Images search for “Malaysia skyscraper,” search results would have shown other buildings, and I would have needed to scroll quite a lot to notice the Telekom Tower. Still, scrolling down would have made sense — don’t give up too quickly!

So Maks was somewhere around Kuala Lumpur. I took to Google Earth Pro, searched for Menara Telekom and turned on the “3D buildings” feature. Thankfully, Kuala Lumpur has only a few buildings visualised that way — including the Telekom Tower. 3D models can help to show what the angle of the photograph is: you can arrange the model in the same way as it is visible in the photo (zoom in, turn left, turn down, etc).

Then I zoomed out and saw green areas (jungle areas) around Kuala Lumpur, especially hilly areas, which looked similar to the area depicted in the photos. (I have to be honest — I forgot to turn on the “terrain” feature, which might have also helped.) In the photograph, you could see other buildings around the Telekom Tower, which was a clue toward letting me know from which direction the photo was taken. The size and angle of the hilly slopes in the photo were also a hint. Looking at satellite photos, I soon noticed a green area close to the city, with visible running trails. Comparing the landscape, the buildings, and the angle of the position, I could more or less precisely determine where Maks was and in which direction he was looking while taking the photo. To confirm I was right, I also searched for photographs of Kuala Lumpur’s running trails and jungle, to see if the flora was the same. And indeed it was. And so I could finally go for my family dinner!

Another photo by Maks that I managed to geolocate was again from an unnamed location, but it turned out still to be Kuala Lumpur.

Again, first I searched for all the visual clues. In this photo, clues included: the building prominently featured in the photo and its structure, a neon sign with an “o” at the end, some kind of (hardly visible) sign on the upper right corner of the building, two foregrounded bundles of cables, the roofing of the building, and something violet on the left. I suspected the “something violet” might be Kuala Lumpur Tower (KL Tower), a known landmark of the city that often glows violet and that also has a roundish shape that seemed to match the object in the photo. The main building looked like a hotel, so I decided to search for hotels close to the KL Tower, from which you could see the landmark. At first I was looking for hotels starting with an “o,” but none of those popping up on Google Maps seemed similar to the hotel in the picture.

Then I went to booking.com and tried to find a hotel with an “o” in the name, but, to be honest, there are so many hotels in Kuala Lumpur, and this method seemed so dull that I gave up! I used Google Street View to walk a bit around the streets of Kuala Lumpur, close to the KL Tower and with some cables also in view. It was still not working, as I couldn’t see any building similar to the building in the picture; plus — there are so many streets in Kuala Lumpur! Then I thought I’d give Youtube a try. Maks’ photo was taken at night, which made it easier to see the typeface for the logo that included the letter “o.” So I searched for videos of drone footage of Kuala Lumpur at night (query: Kuala Lumpur night drone). These videos were quite short, and since I’ve never been to Kuala Lumpur, they actually drew me in and compelled me to watch some more! In the third video I watcged, I noticed the InVito Hotel, with the last letter looking similar to the one on Maks’ photo (the circle of the “o” was slightly broken). Bingo!

I searched for photographs of the hotel, and these seemed to match the building in the photo (the shape of the windows, the top terraces); however, most of the photos did not include the neon sign. Quite possibly I had looked at the hotel earlier on and had excluded it from my research because it had no sign on the top.

I went to Google Street View again and was still quite unconvinced that this building was actually the building. The upper part of it, above the entrance with the roofing, was totally different. I checked for other InVito hotels in Malaysia, but there are none. After a while I realised that the roofing was actually from a different building (I should have noticed this earlier because of the proportions), and the photograph was not showing the hotel’s entrance area, but a side area. So then I looked at the buildings next to the KL Tower, in order to find this second one you could see on the photograph Distinctive features of this building included a small window close to the corner of the building, then white space, and then bigger windows. I did confirm that I was walking (or rather, Google Street View walking) close to the area where Maks was. I zoomed in to see just a bit of the KL Tower and tried to determine the exact location at which the photo was taken. (I’m not 100% sure whether I chose the perfect spot, since only some of the surrounding streets were photographed by Google Street View).

As you can see, for all of that work, I didn’t use any advanced tools — just my brain and just some publicly available, free online applications. And if you wonder why I actually did that, the answer is simple: to have fun, to test and exercise my brain, and to develop a skill that might be very needed at some (or any) point in my professional career. I truly recommend joining the #DigitalSherlocks for those or any other reasons.


Follow along for more in-depth analysis from our #DigitalSherlocks.