|
Home
Curriculum Vitae
Research
Publications
Education
Home at ifp
| |
Brenner, Claus; Boehm, Jan; Guehring, Jens;
| Publication: |
IAPRS, Vol. 32/5, ISPRS, Hakodate, Japan, pp. 209-216 |
| Publication Date: |
04/1998 |
Abstract:
Starting from dense range images that are obtained by a structured light system, a higher-level description is generated. This description is well-suited to perform object recognition and pose estimation.
Once the correspondence between the CAD modell and the measured part is established we are now able to compute the geometric deviation of the manufactured part from ist CAD model.
Brenner, Claus; Boehm, Jan; Guehring, Jens;
AA(Univ. of Stuttgart)
| Publication: |
Proc. SPIE Vol. 3521, p. 237-247, Machine Vision
Systems for Inspection and Metrology VII, Bruce G. Batchelor; John W.
Miller; Susan S. Solomon; Eds. |
| Publication Date: |
10/1998 |
Abstract:
A research group at the University of Stuttgart has set up an experimental
measurement robot for industrial close range inspection. During a test run, the
feasibility of a multi- sensor/actor system has been shown. The system uses
optical sensors to perform different tasks including object recognition,
localization and gauging. It is a step towards systems which are able to inspect
and gauge several parts from a set of parts stored in a 3D model database. This
paper describes the results which have been obtained so far and were
demonstrated during a test run. It then focuses on our latest developments
concerning 3D data acquisition, registration, segmentation, model generation
from CAD data and object recognition.
Brenner, Claus; Boehm, Jan; Guehring, Jens;
AA(Univ. of Stuttgart)
| Publication: |
Proc. SPIE Vol. 3641, p. 164-172, Videometrics VI,
Sabry F. El-Hakim; Armin Gruen; Eds. |
| Publication Date: |
12/1998 |
Abstract:
Presently there is a growing demand for fast and precise 3D computer vision
systems for a wide variety of industrial applications like reverse engineering,
quality control and industrial gauging. One important aspect of any vision
system is the data acquisition. If the principle of triangulation is used the
correspondence problem is to be solved. The coded light approach offers a fast
way to overcome this problem and to provide dense range data. In order to get
high accuracy range images the system needs to be calibrated. In this paper, we
compare two calibration techniques: polynomial depth calibration and
photogrammetric calibration. We have carried out both methods independently. To
obtain results about the accuracy in object space, we measured the surface of a
plane-table.
Boehm, Jan; Brenner, Claus; Guehring, Jens;
| Publication: |
'Mustererkennung 1999', 21. DAGM-Symposium, Bonn,
Germany, Springer, Informatik aktuell, Förstner, W.; Buhmann, J.M.;
Faber, A.; Faber, P., (Hrsg.) |
| Publication Date: |
1999 |
Abstract:
An der Universität Stuttgart wird im Rahmen eines Sonderforschungsbereichs an einem System zur explorativen
Erkennung und Vermessung von Werkstücken geforscht. Ein Ziel der Arbeiten ist der Aufbau eines
Gesamtsystems, bestehend aus Beleuchtung, Aktoren, optischen Sensoren und
Auswertealgorithmen, welches in der Lage ist, ein eingelegtes Werkstück zu
erkennen, dessen Position und Orientierung zu ermitteln und anschließend für dieses Teil
spezifische, vordefinierte Meß- und Prüfaufgaben durchzuführen.Ein Hauptaugenmerk liegt dabei auf der Automatisierung und Robustifizierung der
Objekterkennung, welche den Einsatz des Meßsystems für verschiedene Werkstücke ermöglicht.
In diesem Beitrag zeigen wir die durchgängige Kette von der Modellierung eines Objekts mittels eines 3D-CAD-Systems über die
Fertigung des Teils, die Erfassung und Segmentierung bis hin zur Erkennung dieses Objekts auf.
Boehm; Jan, Brenner, Claus; Guehring, Jens; Fritsch, Dieter
| Publication: |
'IAPRS', Vol. 33/5, ISPRS, Amsterdam, The Netherlands, pp.
76-83 |
| Publication Date: |
2000 |
Abstract:
In this paper we report on our work on a CAD model-based object recognition
system for industrial parts. We present a novel approach which uses information
derived from the CAD model in the early process of range image segmenta-tion.
This approach gives an exact classification of the range image enabling the
development of a CAD based object recognition system. We describe the feature
extraction from CAD data and its use in the curvature based range im-age
classification. We carried out experiments with data from multiple sources. The
results obtained are presented and discussed.
Guehring, Jens; Brenner, Claus; Boehm, Jan; Fritsch, Dieter
| Publication: |
'IAPRS', Vol. 33/5, ISPRS, Amsterdam, The Netherlands, pp.
327-338 |
| Publication Date: |
2000 |
Abstract:
Dense surface acquisition is one of the most challenging tasks for optical 3-D
measurement systems in applications such as inspection of industrial parts,
reverse engineering, digitization of virtual reality models and robot guidance. In
order to achieve high accuracy we need good hardware equipment, as well as
sophisticated data processing methods.
Since measurements should be feasible under real world conditions, accuracy,
reliability and consistency are a major issue.
Based on the experiences with a previous system, a detailed analysis of the
performance was carried out, leading to a new hardware setup. On the software
side, we improved our calibration procedure and replaced the phase shift
technique previously used by a new processing scheme which we call line shift
processing.
This paper describes our new approach. Results are presented and compared to
results derived from the previous system.
Boehm, Jan; Guehring, Jens; Brenner, Claus
| Publication: |
Proc. SPIE Vol. 3409, Videometrics VII,
Sabry F. El-Hakim; Armin Gruen; Eds. |
| Publication Date: |
2001 |
Abstract:
We present our work on the mplementat on and cal brat on of a mult sensor measur
ng system.The work s part of a large scale research project on opt cal
measurement us ng sensor actuator coupl ng and act ve explorat on.Th s project
is a collaboration of researchers from seven institutes of the University of
Stuttgart ncluding photogrammetry, mechanical eng neering and computer
science.The system consists of optical sensors which can be manipulated in
position and or entation by robot actuators,and light sources which control
illumination.The system performs di .erent tasks ncluding object
recognition,localization and gaug ng.Flexibility s achieved by replacing the
common serial measurement cha n by nested control loops nvolv ng autonomous
agents wh ch perform basic tasks n a modular fash on.The system s able to nspect
and gauge several parts from a set of parts stored n a 3-D model database. The
paper gives an overview of the ent re system and details some of the
photogrammetry-related aspects such as the cal brat on of the d .erent sensors (cameras,stereo-head,str
pe projector),the cal brat on of the measurement robot using photogrammetric
measurements,as well as data processing steps like segmentation,object pose
determination, and gauging.
Guehring, Jens
| Publication: |
Proc. SPIE Vol. 3409, Videometrics VII,
Sabry F. El-Hakim; Armin Gruen; Eds. |
| Publication Date: |
2001 |
Abstract:
This paper describes the implementation of a LCD stripe projection system, based
on a new processing scheme called line shift processing. One advantage of the
method is, that the projection device can, but is not required to be stable over
time nor does it have to be calibrated. The new method therefore allows us to
use an off-the-shelf multimedia projector to acquire dense 3-D data (several
100K Triangles) of large objects in a matter of seconds. Since we are able to
control our source of illumination, we can also acquire registered color
information using standard monochrome cameras. Besides the description of the
system some practical examples are given. The calibration of the system is also
addressed.
Guehring, Jens
| Publication: |
Third International Conference on 3D Digital Imaging
and Modeling (3DIM'01), Québec City, Canada, IEEE |
| Publication Date: |
2001 |
Abstract:
We present a complete data acquisition and processing chain for the reliable
inspection of industrial parts considering anisotropic noise. Data acquisition
is performed with a stripe projection system that was modelled and calibrated
using photogrammetric techniques. Covariance matrices are attached individually
to points during 3D coordinate computation. Different datasets are registered
using a new multi-view registration technique. In the validation step, the
registered datasets are compared with the CAD model to verify that the measured
part meets its specification. While previous methods have only considered the
geometrical discrepancies between the sensed part and its CAD model, we also
consider statistical information to decide whether the differences are
significant.
|
