Organisation/Company: AIT Austrian Institute of Technology GmbH
Research Field: Computer science Mathematics Physics
Researcher Profile: First Stage Researcher (R1)
Application Deadline: 19/10/2022 21:00 - Europe/Brussels
Location: Austria › Vienna
Type Of Contract: Temporary
Job Status: Part-time
Hours Per Week: 20
We are Austria's largest research and technology organisation and an
international player in applied research for innovative infrastructure
solutions. This makes us a powerful development partner for industry and a top
employer in the scientific community. Our Center for Vision, Automation &
Control in Vienna invites applications for a:
Master Thesis “Implementation of a combined multi-view laser line 3D
- Center for Vision, Automation & Control -
The inline computational imaging (ICI) technique of our Competence Unit
High-Performance Vision Systems (HVS) is based on a single-image
sensor technology including corresponding algorithms for image
reconstruction, for simultaneous 2D and 3D quality inspection of
manufactured parts. Our ICI pipeline reconstructs the acquired object in
3D by applying a multi-view stereo algorithm, by finding the same
features/patterns in multiple images. This is only possible if the object
exhibits some surface texture, such as banknotes, wafers, or steel billets
for instance. This is not the case for objects that are rather smooth,
such as keys for instance, at spatial resolutions at which they are
On this account, this master's thesis project will take a step towards a
new “Laser-augmented ICI”, which extends the current standard setup for
ICI by an additional laser line illumination. In presence of an object the
laser line is bent by the object's surface to various degrees (as seen by
the camera's sensor), depending on the local surface curvature. This
distortion of the initially straight laser line can be used to infer
spatially resolved depth information of the surface, even for objects
without inherent surface texture. Including this extra depth cue into the
reconstruction algorithms of the ICI pipeline would allow the inspection
of objects more reliably, which exhibit no significant inherent surface
texture at relevant length scales for the inspection task.
The main goal of this master's thesis is to investigate and evaluate the
potential improvement of the 3D reconstruction quality, that can be gained
with a new “Laser-augmented ICI” compared to standard ICI. The working
hypothesis to be addressed in this work is whether additional laser line
illumination and derived depth cues thereof improve the accuracy of the 3D
reconstruction inferred from the acquired data. The basis of this work is
a portable ICI setup that has been developed within a previous master's
To this end, the following specific aims are proposed:
(1) Literature research of the state-of-the art of laser line triangulation methods. This results in:
Design and specifications of the extended “Laser-augmented ICI” setup (i.e., additional laser line illumination).
A selection of candidate laser line-based 3D reconstruction algorithms that infer depth information by analysing the distortion of a laser line.
(2) Extension of an available ICI setup to “Laser ICI”. This includes:
Adjusting an available ICI setup according to the designs/specifications from specific aim (1).
Acquiring laser line image data that is registered with the ICI image stacks.
Adapting and/or implementing candidate laser line-based 3D reconstruction algorithms identified in specific aim (1).
3D reconstructing of the acquired data laser line-based 3D reconstruction algorithms. Ideally, this will be extended to a combined approach that uses the multi-view stereo approach (e.g., standard ICI) and laser-based depth cues at the same time.
(3) Evaluation of the selected laser-based 3D reconstruction approaches, by:
Defining and implementing figure of merits to evaluate the accuracy of 3D measurements from standard ICI and “Laser-augmented ICI” data.
Collecting image data for ground truth test objects, i.e., objects with known geometry.
Comparing the (added) benefit of the laser-based 3D reconstruction approaches with each other.
Benchmarking 3D reconstruction results of the selected test objects from “Standard ICI” and “Laser-augmented ICI”.
Your project involves hands-on lab work, namely image acquisition of data
sets and extension/adjustment of available imaging setups.
Furthermore, it entails practical programming work, i.e., the adaptation
and/or implementation of laser-based 3D reconstruction algorithms and
existing image acquisition software.
Finally, there will be testing of those algorithms and evaluating the
resulting 3D reconstructions to assess the performance and added benefit
of laser line illumination.
We expect that laser illumination and its additional depth cues will
benefit the 3D reconstruction for objects with marginal inherent surface
texture. The multi-view stereo matching algorithm, which is a core element
of our ICI 3D reconstruction pipeline, requires surface texture to work
reliably. In theory, the additional laser-based depth cue should be
independent of the object's surface texture, and hence, is expected to
improve the 3D reconstruction quality.
You will get familiar with state-of-the-art 3D reconstruction methods that
are at the forefront of application-driven research in the field.
You have the unique opportunity to work on an interesting real-world
problem that arose in ongoing collaborations with industrial partners. The
Competence Unit High-Performance Vision Systems (HVS) at AIT has been
active in research for industrial inspection and quality assurance systems
for over 20 years, which creates a stimulating and supportive framework
for the students working with us.
Your qualifications as an Ingenious Partner :
Ongoing Master's studies in Computer Vision/Image Processing, Computer
Science, Physics, Mathematics or similar
Hands-on lab work experience with image acquisition using
industrial/digital imaging is advantageous
Programming skills in Python and C++ are expected
Strong skills in image processing are expected
Knowledge about Blender are advantageous
Interest in technology development and applied research driven by real-
world problems is expected
Good communications skills and ability to work in a multidisciplinary team
Very good command of English (in writing) is expected
What to expect:
EUR 828,-- gross per month for 20 hours/week based on the collective
agreement. There will be additional company benefits. You will be part of our
international Young AIT network. As a research institution, we are familiar
with the supervision and execution of master' theses, and we are looking
forward to supporting you accordingly
At AIT, the promotion of women is important to us - that's why we are
especially looking forward to applications from female students!
Please submit your application documents including CV, cover letter and
certificates (transcript of records and other relevant certificates) online.
TOMORROW TODAY - WITH YOU?
Organisation/Company: AIT Austrian Institute of Technology GmbH
Department: Human Resources
Organisation Type: Public Research Institution
Website: https:// www. ait.ac.at
Postal Code: 1210
Street: Giefinggasse 4