Mobile Game Tracking Rig
Posted: Sat Aug 17, 2019 4:53 pm
A friend likes to play a mobile phone game (no names). One of the functions is to physically visit locations/cells to capture them These cells are around 300m in diameter, so there are lots of them around. Unfortunately as of August 2019, there is no method of viewing the places you have visited. While he is notified if someone captures one of his cells, it only says how many kilometres away it is. Not very handy of knowing really on where this might have been located.
Since we have been driving to Perth and back a bit lately and seeing the game being played out, I thought it might be fun to try and build a data capture rig. This would:
1) capture a screenshot of the mobile phone game for each cell visited
2) OCR the captured screen
3) store all the data found, ie: name of cell, Latitude/Longitude location, Cell Owner, Cell income etc
4) record other stats and data via keyboard hot-keys: Mobile phone towers locations (handy to know where good signal reception is), in game monsters, special buildings etc
5) record stats of the journey
6) other hot-keys: Audio on/off, servo on/off, pause data capture, extra screenshots
The hardware for this rig:
* Raspberry Pi 2B
* Raspberry Pi Camera
* HDMI Pi display (was going to use a RPi touch display)
* Adafruit Ultimate GPS 3 module
* Two 9Ahr 12V LCA batteries (I had them lying around)
* buzzer (audio feed back)
The initial idea was to capture cell location data, cell name and owner etc. The project has kind of grown from there and perhaps way too much time has been spent on this whimsical project, but it has been a lot of fun?
The software is written in the wonderful Pygame python: https://www.pygame.org/news and uses various imported modules: gpiozero, pytesseract (does the OCRing), picamera, PIL (image manipulation), GPSController and bunch of normal ones, ie: maths, time etc.
Here is a photo of it working out on a drive:
We quickly figured out it needed a sun shade to keep the sun off of the mobile phone screen. OCR doesn't like reflections! It's hard to see the screen, so, here is a close up:
Once we have recorded a journey/drive, the data can be imported into QGIS: https://qgis.org/en/site/, so our GPS data can be nicely viewed: Above, I have included the Telstra mobile phone tower locations that we have recorded.
Since we have been driving to Perth and back a bit lately and seeing the game being played out, I thought it might be fun to try and build a data capture rig. This would:
1) capture a screenshot of the mobile phone game for each cell visited
2) OCR the captured screen
3) store all the data found, ie: name of cell, Latitude/Longitude location, Cell Owner, Cell income etc
4) record other stats and data via keyboard hot-keys: Mobile phone towers locations (handy to know where good signal reception is), in game monsters, special buildings etc
5) record stats of the journey
6) other hot-keys: Audio on/off, servo on/off, pause data capture, extra screenshots
The hardware for this rig:
* Raspberry Pi 2B
* Raspberry Pi Camera
* HDMI Pi display (was going to use a RPi touch display)
* Adafruit Ultimate GPS 3 module
* Two 9Ahr 12V LCA batteries (I had them lying around)
* buzzer (audio feed back)
The initial idea was to capture cell location data, cell name and owner etc. The project has kind of grown from there and perhaps way too much time has been spent on this whimsical project, but it has been a lot of fun?
The software is written in the wonderful Pygame python: https://www.pygame.org/news and uses various imported modules: gpiozero, pytesseract (does the OCRing), picamera, PIL (image manipulation), GPSController and bunch of normal ones, ie: maths, time etc.
Here is a photo of it working out on a drive:
We quickly figured out it needed a sun shade to keep the sun off of the mobile phone screen. OCR doesn't like reflections! It's hard to see the screen, so, here is a close up:
Once we have recorded a journey/drive, the data can be imported into QGIS: https://qgis.org/en/site/, so our GPS data can be nicely viewed: Above, I have included the Telstra mobile phone tower locations that we have recorded.