PRC - Pattern Recognition Company

Human Vision and Electronic Imaging XVII

Fri, 20 Jan 2012 10:02:00 +0100

We are pleased to announce that we participate in the conference "Human Vision and Electronic Imaging XVII", which takes place from 23-26 January in San Francisco, USA. There we present current results of our project "Aktiv Bild", where we cooperate with Nik Software and which is funded by the German Federal Ministry of Education and Research (BMB+F) within the scope of the program KMU-innovative.

You are welcome to join us in San Francisco. For further information do not hesitate to contact us.

Nik Software and PRC start cooperative Project "AktivBild"

Tue, 30 Nov 2010 09:50:00 +0100

Due to their cooperation in the scope of computational photography Nik Software GmbH and PRC GmbH are happy to announce the beginning of the project "AktivBild". Within this project algorithms of information technology and innovative software will be developed, that offer new ways of image acquisition and processing. Thereby taking a series of images, image fusion and image quality assessment play a key role. Consequently next generation of digital photography not only copies its analog counterpart, but by combining techniques from pattern recognition, image processing and machine learning digital photography will be redefined.

The project "AktivBild" is funded by the the German Federal Ministry of Education and Research (BMB+F) within the scope of the program KMU-innovative.

Collaboration with Raytrix on 4D light-field cameras

Mon, 01 Mar 2010 19:28:00 +0100

Raytrix GmbH Kiel and PRC announced to collaborate in the area of light-field cameras. Raytrix is the first company worldwide to offer light-field cameras, a new kind of imaging device with

up to six times the field of depth of a standard camera at same aperture and resolution

the highest ever achieved effective resolution ratio for a plenoptic camera

real time refocusing on arbitrary focal surface from a single snapshot.

The cameras developed and marketed by Raytrix open new opportunities for optical inspection and digital photography, a potential that is now enhanced by PRC´s pattern recognition expertise.

PRC technology featured in Springer journal

Mon, 01 Feb 2010 13:00:00 +0100

We are pleased to announce that our research article "Optical Inspection of Welding Seams" is published in the Springer series Communications in Computer Science and Information Science, Vol. 68, pages 269 - 282, 2010.

Abstract

We present a framework for automatic inspection of welding seams based on specular reflections. To this end, we make use of a feature set – called specularity features (SPECs) – that describes statistical properties of specular reflections. For the classification we use a one-class support-vector approach. We show that the SPECs significantly outperform other approaches since they capture more complex characteristics and dependencies of shape and geometry. We obtain an error rate of 3.8%, which corresponds to the level of human performance.

PRC titled as global market leader

Mon, 01 Feb 2010 12:00:00 +0100

VDI Nachrichten announced PRC as one of Germany's world market leaders of the R&D section "custom made solutions and software for image processing and pattern recognition". To have a look at Germany's world market leaders visit http://www.vdi-nachrichten.com.

PRC helps sustain the ARTTS project

Tue, 12 Jan 2010 12:00:00 +0100

The University of Lübeck and PRC will collaborate to ensure the sustainability of the ARTTS project. ARTTS has been a very succesfull European project to develop the world’s smallest ToF camera and world’s first and only HDTV+ToF camera. Moreover the focus of ARTTS was on algorithms that improve the cameras and enable applications. Based on the results of the ARTTS project and its own developments, PRC is now offering software modules and systems for ToF based applications. Further information about the ARTTS project can be found at www.artts.eu.

Collaboration with Nik Software on Computational Photography

Fri, 01 Jan 2010 12:00:00 +0100

This is a strategic partnership that will shape the future of digital imaging.
Manuel Wille, Vice President Engineering, Nik Software

Nik Software, a global player in the digital-imaging and photography market, and PRC announced to collaborate in the area of advanced image processing and pattern recognition with a focus on Computational Photography. The collaboration will extend the range of innovative technologies that make digital imaging and photography easier and more powerful. „This is a strategic partnership that will shape the future of digital imaging.“ said Manuel Wille, Vice President Engineering at Nik Software.

New Company Flyer Released

Thu, 15 Oct 2009 12:00:00 +0200

Vision 2009 in Stuttgart

Sun, 11 Oct 2009 00:00:00 +0200

At booth 4C73, we presented some of our latest projects, products and our expertise in image processing and machine learning tasks. Our large-scale scanning device with an optical resolution of up to 4 micrometers with included customizable defect detection was a great attraction as well as our custom-made image enhancement software for removing illumination inhomogeneity from images. Finally, we recognized enormous interest in the Time-of-Flight technology. In fact, we are one of the first companies to offer expertise and algorithms for industrial applications based on this novel technology.

PRC technology featured in Inspect journal

Thu, 01 Oct 2009 00:00:00 +0200

We are pleased to announce that our article The Needle in the Haystack is published in the Inspect journal, 10, pages 74 - 76, 2009.

PRC has received R&D grant

Sat, 01 Nov 2008 00:00:00 +0100

The project Maledo (Machine Learning for Self-Diagnosis and Self-Optimisation) was initiated to develop novel systems for optimizing production processes and for triggering smart alarms in case of system or sensor failures. Based on state-of-the art machine learning techniques, we provide a framework for the diagnosis of critical system states. Applications range from process control in the chemical industry over steel production processes to patient monitoring in intensive care units.

In contrast to conventional approaches, we do not analyse each sensory input separatly but combine the output of multiple sensors to model the multivariate inter-dependencies. Thus, the diagnosis performance is significantly increased while reducing the false alarm rate. In collaboration with our partner iba AG we even go a step further and develop strategies which allow to optimize the efficency and quality of a production line based on smart alarm systems. Funding is provided by the BMBF, the German Federal Ministry of Education and Research.