The Signal Analysis and Machine Perception Laboratory (SAMPL) was founded in 1986 by Prof.
Kim L. Boyer to undertake research in the broad areas of computer vision and image
analysis. While the specific topics under study at any given time may vary according to current sponsorship and faculty and student interests, SAMPL has a rich, productive history of research and publications. The following application areas
have been, or are currently, foci of our program:
 | Aerial and satellite image understanding, including hyper- and
multispectral imagery, for automated mapping, detection of human activity,
and change detection using photometric features and textures, and perceptual
organization |
| Retinal boundary detection and nerve head tracking in Optical Coherence
Tomography |
| Automated analysis of magnetic resonance imagery of the head and, more
recently, the heart |
| Range image and terrain elevation data segmentation and analysis based on
surface curvatures |
| 3D object recognition using graph-theoretic approaches, appearance-based
methods (both local and global features) |
| Perceptual organization for surface and contour decomposition from
curvature measures |
| Range profile data analysis for target detection and classification in
heavily occluded environments |
| Head and face tracking for driver behavioral analysis |
| Human motion understanding, including gait analysis |
| Automated range image registration for efficient modelbuilding |
| Automated weld pool analysis for on-line quality control in robotic
welding |
| Real time fillet weld inspection from range data |
| Robust methods in surface organization; pipeline corrosion detection |
Within these applications, we have always endeavored to produce high-quality,
rigorous, fundamental research results that find application across a broad
swath of computer vision and image processing problem domains.
Sponsors
We are a leading research group, with a strong, respected presence in the
international computer vision research community. We have placed our alumni in
academic and industrial positions across the US and around the world. Our
sponsors and collaborators over the years have included:
SAMPL researchers were proud to receive the Siemens Best Paper Award at the
1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
(CVPR) in New York City. We maintain close working ties with our colleagues in
the Electrical
and Computer Engineering
department at OSU, as well as those within the
Computer
Science and Engineering department.
Facilities
Owing in part to a recent National Science
Foundation equipment grant, SAMPL is one of the best equipped academic
laboratories for research in computer vision in the United States.
SAMPL is physically organized into two adjoining rooms. The main room
encompasses approximately 770 square feet (73 m2) and includes
seven high performance (Pentium 4 @ 1.7 GHz) image processing workstations
(3 with 17" dual monitors and 4 with 21" monitors) running Windows
2000, one HP Visualize X Class workstation with HP fx10 graphics card,
wireless stereo viewing glasses and 1.25 GB RAM,
meeting and presentation area with Sharp
Notevision computer projection system, stereo television and VCR, and an
informal discussion area.
The second, or sensors, room encompasses 396 square feet (38 m2)
and includes a Minolta Vivid high
performance laser ranging system with computer controlled turntable and
dedicated workstation, many digital cameras of various types, GL1 and ZR10 Canon
digital camcorders with fire-wire connections, floor-to-ceiling gantry
for the mounting of camera and lighting fixtures, controlled spot and
flood lighting for the imaging bay,14 cpu Beowulf
commodity supercomputer and a mix of
high-performance image processing workstations, multiple high-capacity network server appliances (nearly 1
terabyte of online storage), dual cpu compute server, Matrox 4-sight image
acquisition and processing unit with hardware compression board installed, Point
Grey Triclops color stereo imaging system with dedicated workstation, HP
color and monochrome laser printers, galley, lab library, and equipment storage.
|
