Road To Nowhere — Geolocating a village in Myanmar from satellite imagery

Solving this Hacktoria challenge the hard way.

Mission Briefing:

Special Agent K, we’ll be needing your assistance on a geolocation matter. One of our clients, a medium sized European goverment agency, has requested we find a needle in a haystack.

Our client is involved in fighting terrorist groups in Asia and Africa. Recently, they gained access to a Dropbox account filled with map data. Most of these seem to be related to safe houses used by a terrorist organization called “The Meercats”. Indeed a strange name, but nontheless, they mean business.

After having located most safehouses, our client is still struggling to locate a few of them. Since the operation to take down all of these locations needs to be synchronized. There’s an urgent need to identify them all before the next phase can begin.

Below you find an image of what appears to be a village, with a red marker to indicate the exact location. It’s your assignment to find where in the world this is. Currently we have reason to believe this is in Africa or Asia. But, given the organizations’ widespread members, this is only a guess at this time.

Use the coordinates you find as the password to open the ZIP archive. Inside, you find the linkfile to your Contract Card as always. Also, we heard the number 1920 is of importance in this case…

As always. Special Agent K, the contract is yours, if you choose to accept.

I must say, when I fist saw this image I was a little intimidated, since it’s not your everyday CTF geolocation. Especially because, as it turns out, there’s a way simpler solution that involves steganography, but I like to do things the hard way.

Let’s start by searching for some clues in the description.

The ‘terrorist organization’ cited in the brief is called ‘The Meercats’, which is an animal (spelled ‘meerkat’) that lives in southern Africa. This aligns with the belief that the village might be in Africa.

But, after looking at some villages around southern and center Africa, I realized that they don’t really look like the one in the image, so I shifted towards Asia.
What I mean by that is, taking a look at the color of the terrain and the distribution of the trees compared to the image of the village, they don’t really match.

Right now, instead of looking at specific details, it’s best to focus on the larger picture.
Our aim at this stage is to find a country (or even a part of a continent) that roughly matches our image.

For example, before examining Asia I was looking trough Africa. But Africa is a big, big continent, so how do we search for a small village?

First, we exclude all the areas that cannot match with our image.
In this map, I excluded the areas inside the red box since they are way too dry to match our image.
The blue box area is a good place to start, since it could match our terrain.

All I did at this stage was zoom in on random villages and compare them with the image, jumping on different places inside the blue box.
But they don’t really match, so I discarded the whole continent and moved to Asia.

Satellite image layered with a topographical map, since we are looking for a slightly mountainous zone.

After some time looking at various parts of Asia, I decided to search Myanmar in depth, because the villages and the levels of green look similar. Especially, something very usefull, many roofs are painted blue like in the village we are looking for.

But how did I get to Myanmar? Well, same process as for Africa. Some countries, like Thailand or Malesia, look too green. India looks interesting, in fact I spent some time looking there, but the villages didn’t match.

Unfortunately, when you don’t have many clues at the start, you have to spend some time jumping from place to place and sort what does and doesn’t match.

Now that we have a clue as what country the village might be in, let’s lay out the major features we are looking for:

1. The lake on the top center of the image.
2. The trees on the left of the picture, that follow a very interesting shape, most likely caused by the proximity with a mountain.
3. The road that goes trough the village.
4. The very red road on the down-right corner. Might be an artefact from the satellite imagery, but it’s worth noting.

Taking all those clues, we can start searching.

After some time spent on the southern Myanmar, I moved towards the capital as it’s much more similar. Emphasis on point 2.
In this part of the world, I find that using maps from HERE WeGo is much more useful than anything Google in this case.

Let’s select the areas we should take into consideration.

First, we can exclude the northem and western areas, as they are way too mountainous. The ‘tail’ doesn’t match either, so let’s focus on the center of the country, especially that big spot that’s flatter and dryer.

The yellow spot in the center of the country is where we should search.

Now I was completely sure that this area was the right one, so here is where the real work starts: finding that exact village.
Something that helps here is a topographic map, because of point 2.

Let’s move towards a more bumpy terrain, like around Mount Popa.

At this point, all we have to do is search. This is the part where you just go village to village looking for the right one, or to find some clue that could lead to it, like what happened here.
After a lot of searching, I found the reddish road that I was referring to on point 4. Following this road, finally, leads us to the village we were looking for.

To get the exact point on the marker, just look at the house with the blue roof under it and the road on the right.

Pretty easy to do it by eye.

Coordinates: 20.899370,95.118041
We also have to specify the zoom level for some reason, which is 16.
To get both using HERE WeGo, look at the URL.

So using ‘20.899370,95.118041,16’ as our password we can extract the file and get the contract card.