TRAFFIC LIGHTS DETECTION
​
A Deep Learning model has been developed which, by means of the image captured by the Py Camera, detected the traffic lights, their distance to the camera and their status (green, amber or red), so that depending on the distance and their status, the speed of the scooter was limited according to the requirements of the test. We used the BOSCH Dataset.
In addition, a Dashboard was also created to control the execution of the developed program.
​
​
​
​
​
​
​
​
​
​
​
Visit the Smart Scooter Etseib repository for more technical information.
 |  |
|---|
CONES DETECTION
​
A Deep Learning model has been developed which, by means of the image captured by the camera, detects the cones and their distance from the camera and, depending on their positions, limits the speed of the scooter according to the requirements of the test. In addition, a Dashboard was created to control the execution of the developed program.
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
Visit the Smart Scooter Etseib repository for more technical information.
 |  |
|---|
LANE DETECTION
​
The objective of this part of the development is to, by means of computer vision algorithms, detect driving lane lines, as well as, according to the captured inclination of the lines, define whether the scooter is in or out of the lane.
​
​
​
​
​
​
​
​
​
Visit the Smart Scooter Etseib repository for more technical information.
