Spotlight on: Stereovision
Creative technical knowledge becomes more important to stand out in this ever-evolving digital world. We aim to acquire the best talent, even those still graduating. That is why we offer the opportunity for students of different professions and specialisms to graduate with a thesis at Avular. We are proud to have the best talents joining our team. After graduating, some of them stay at Avular, some continue their studies and some go travel the world. Just before the end of their project, we ask them to write an article about their graduation thesis.
This week, the spotlight is on Petros Kavvathas. He recently graduated from Electrical Engineering at Hogeschool Utrecht with his thesis about stereovision.
What is stereovision?
“Stereovision is a method to extract 3D information from digital images. Specifically, a typical stereovision setup consists of two cameras. The images taken by these two cameras are compared and analyzed to obtain a 3D depth image. This is, in essence, a 3D map of the world from the camera’s point of view, but only for those points that the stereovision setup can see with both cameras. This is very comparable to our eyes, since we also need both eyes to see depth. Avular aims at developing a set for 3D stereovision solutions for robot navigation purposes, and this graduation project has laid out the basis for further development at Avular.
Going to the next level with Stereovision
I developed the first Stereovision camera setup of Avular. Stereovision will be used to navigate through the environment without the need of additional infrastructure, such as beacons placed in the world. Using stereovision does not only reduce the time to deploy a robot, but it is also a very versatile system that can be used in many different environments and scenarios.
However, the wide applicability of stereovision also poses a challenge. This applicability can be realized through all the parameters that you can set yourself of a stereovision setup, that should be optimized for a specific application. The goal is to develop a stereovision setup that can be used directly in a broad range of applications. This is why modularity has been one of the key design elements of this project. The user can change the lenses for a different field of view, determine the distance between the cameras by moving the image sensors further apart and pivot the imaging sensors in order to test with one or both cameras tilted, which is a non-standard setup. In this way, many different configurations of the stereovision camera can be tested out, and the whole setup is useful in a wide range of applications.
This Stereovision camera will be used for prototyping different real-world scenarios, in which the exact application does determine specific requirements for the camera. In order to get the best results available, we chose to use sensors that outperform all sensors used in competing products on the market. In this way, the performance can always be limited to see the effect of using less high-end hardware in specific situations.
This camera will help Avular to analyse all the necessary requirements and possibilities of stereovision. It offers an enormous amount of flexibility, and ensures that every potential stereovision setup can be tested. It will form the basis for successful stereovision, and robots roaming around our world freely.
A personal note
The time I’ve been at Avular has been amazing. Even though my project has been on a new and complex level for me, using the knowledge of my colleagues at Avular I managed realize a product that Avular can actually use to further their company.”