Image Processing Research

Theses

• Bachelor/Master/Diploma Theses Topics

Deformation Models

The Euclidean distance has been successfully used e.g. in optical character and object recognition and has been extended by different methods. As the Euclidean distance does not account for any image transformation (such as the affine transformations scaling, translation and rotation) if they are not part of the training corpus, the tangent distance (introduced by [Simard & Le Cun+98]) as described in [Keysers & Macherey+ 01] is one approach to incorporate invariance with respect to certain transformations into a classification system. Here, invariant means that image transformations that do not change the class of the image should not have a large impact on the distance between the images. A more detailed description can be found in [Keysers & Macherey+ 04] and application in medical image classification can be found in [Keysers & Dahmen+ 03].

The image distortion model has been examined at the Lehrstuhl für Informatik 6 of the RWTH Aachen [Keysers & Dahmen+ 03] and further research is presented in [Keysers & Gollan 03]. The image distortion model is an easily implemented method allowing for small local deformations of an image.

For further information on this topic:

• W2D Software, for appearance based image recognition supporting different pixel to pixel deformation models
• Tangent Distance

Content-based image retrieval

FIRE (Flexible Image Retrieval Engine) is our content-based image retrieval system. Content-based Image Retrieval is an active field of research and is getting more and more important by the rise of personal digital cameras.

A large problem in content-based image retrieval is performance evaluation. Recently there are some efforts on this topic:

• ImageCLEF
• Benchathlon

In our CBIR system FIRE, we have investigated different features, distance functions and ways to measure the retrieval performance of CBIR in different applications, e.g. image retrieval of medical radiographs.

FIRE is available under the terms of the GNU General Public license. Further information on FIRE, the source code, and relevant publication can be found here.

An online demonstration of FIRE is available here. The server usually is running. If it is not up, don't hesitate to complain about it. I am happy to make it run again.

Object recognition

Recognition of objects in cluttered scenes is an interesting and important open problem in computer vision and pattern recognition. In close connection to the development of our content-based image retrieval system FIRE we are focussing on this topic.

In contrast to our earlier approaches were the object were modelled holistically now we follow the promising approach of modeling objects as a collection of parts. This approach offers some immediate advantages:

• changes in the relation between parts can be modeled
• occlusion can be handled very well

Currently we are investigating in combinations of discriminative and generative models and in using spatial relationship between the parts.

Some efforts of out work in this area are documented by our CVPR 05 publication and by our results in the PASCAL Visual Object Classes Challenge 2005.

Most of our software for this work is part of the FIRE framework.

Gesture and Sign Language Recognition

Developing sign language applications for deaf people can be very important, as many of them, being not able to speak a language, are also not able to read or write a spoken language. Ideally, a translation systems would make it possible to communicate with deaf people. Compared to speech commands, hand gestures are advantageous in noisy environments, in situations where speech commands would be disturbing, as well as for communicating quantitative information and spatial relationships.

A gesture is a form of non-verbal communication made with a part of the body and used instead of verbal communication (or in combination with it). Most people use gestures and body language in addition to words when they speak. A sign language is a language which uses gestures instead of sound to convey meaning combining hand-shapes, orientation and movement of the hands, arms or body, facial expressions and lip-patterns. Contrary to popular belief, sign language is not international. As with spoken languages, these vary from region to region. They are not completely based on the spoken language in the country of origin.

Sign language is a visual language and consists of 3 major components:

• finger-spelling: used to spell words letter by letter
• word level sign vocabulary: used for the majority of communication
• non-manual features: facial expressions and tongue, mouth and body position

A more detailed description can be found in [Dreuw & Rybach+ 07]

For further information on this topic:

• website of Philippe Dreuw
• current issues in Sign Language research
• Benchmark databases for video-based automatic sign language recognition and translation.

RWTH-OCR - Offline Handwriting Recognition

The RWTH-OCR system is based on the open-source speech recognition framework RWTH-ASR - The RWTH Aachen University Speech Recognition System, which has been extended by video and image processing methods.

The system has been used for Arabic, Farsi, English, Spanish, and French handwritten texts.

Arabic handwriting recognition -- Due to Parts of Arabic Words (PAWs), white space models and low loop transitions are important in Arabic handwriting recognition.

The visualization shows a training alignment of an Arabic word to its corresponding HMM states, trained with an HMM based system. We use R-G-B background colors for the 0-1-2 HMM states, respectively, from right-to-left. The position-dependent character model names are written in the upper line, where the white-space models are annotated by 'si' for 'silence'; the state numbers are written in the bottom line. Thus, HMM state-loops and state-transitions are represented by no-color-changes and color-changes, respectively.

A more detailed description can be found in White-Space Models for Offline Arabic Handwriting Recognition (ICPR 2008). Extensions using Writer Adaptive Training (WAT) and modified MMI training for model adaptation have been presented in Writer Adaptive Training and Writing Variant Model Refinement for Offline Arabic Handwriting Recognition (ICDAR 2009) and Confidence-Based Discriminative Training for Model Adaptation in Offline Arabic Handwriting Recognition (ICDAR 2009).

Medical Image Processing and Retrieval

Being members of the IRMA project we apply many of our techniques to medical images. We have been successful in medical image retrieval:

• We have been very successful in the medical retrieval task of ImageCLEF 2004.
• We have been very successful in the automatic annotation task of ImageCLEF 2005.

Publications of the Image Processing Group

For further publications you can either

• have a look at the publications of the chair
• or look at the members personal websites for most up-to-date information

Selected Publications

2020	2019	2018	2017	2016	2015	2014	2013	2012	2011	2010	2009	2008	2007	2006
2005	2004	2003	2002	2001	2000	1999	1998	1997

Page: 1

• D. Keysers, F. J. Och, and H. Ney. Efficient Maximum Entropy Training for Statistical Object Recognition. In Informatiktage der Gesellschaft für Informatik, pages 342-345, Bad Schussenried, Germany, November 2002.
  
• B. Haasdonk, and D. Keysers. Tangent Distance Kernels for Support Vector Machines. In International Conference on Pattern Recognition (ICPR), pages 864-868, Quebec City, Canada, September 2002.
  
• D. Keysers, F. J. Och, and H. Ney. Maximum Entropy and Gaussian Models for Image Object Recognition. In Deutsche Arbeitsgemeinschaft für Mustererkennung Symposium (DAGM), pages 498-506, Zürich, Switzerland, September 2002.
  
• T. M. Lehmann, M. O. Güld, C. Thies, K. Spitzer, B. Wein, M. Kohnen, H. Schubert, R. Günther, D. Keysers, and H. Ney. Inhaltsbasierter Zugriff auf große medizinische Bilddatenbanken. Informatik, Biometrie und Epidemiologie in Medizin und Biologie, volume 33, number 2--3, pages 163-164, September 2002.
  
• R. Paredes, E. Vidal, and D. Keysers. An Evaluation of the WPE Algorithm using Tangent Distance. In International Conference on Pattern Recognition (ICPR), pages 48-51, Quebec City, Canada, September 2002.
  
• S. Molau, F. Hilger, D. Keysers, and H. Ney. Enhanced Histogram Normalization in the Acoustic Feature Space. In International Conference on Spoken Language Processing (ICSLP), volume 1, pages 1421-1424, Denver, CO, USA, September 2002.
  
• D. Keysers, R. Paredes, H. Ney, and E. Vidal. Combination of Tangent Vectors and Local Representations for Handwritten Digit Recognition. In Int. Workshop on Statistical Pattern Recognition (SPR), pages 538-547, Windsor, Ontario, Canada, August 2002.
  
• D. Keysers, R. Paredes, H. Ney, and E. Vidal. Combination of Tangent Vectors and Local Representations for Handwritten Digit Recognition. In Int. Workshop on Statistical Pattern Recognition (SPR), Lecture Notes in Computer Science, pages 538-547, Windsor, Ontario, Canada, August 2002.
  
• B. Wein, T. M. Lehmann, D. Keysers, H. Schubert, and M. Kohnen. Detailed Image Classification Code for Image Retrieval of Medical Images (IRMA). In Int. Congress and Exhibition on Computer Assisted Radiology and Surgery (CARS), pages 513-517, Paris, France, June 2002.
  
• M. O. Güld, B. Wein, D. Keysers, C. Thies, M. Kohnen, H. Schubert, and T. M. Lehmann. A Distributed Architecture for Content-Based Image Retrieval in Medical Applications. In International Workshop on Pattern Recognition in Information Systems (PRIS), pages 299-314, Alicante, Spain, April 2002.
  
• R. Paredes, D. Keysers, T. M. Lehmann, B. B. Wein, H. Ney, and E. Vidal. Classification of Medical Images using Local Representations. In Bildverarbeitung für die Medizin (BVM), pages 171-174, Leipzig, Germany, March 2002.
  
• D. Keysers, S. Celik, H. Braess, J. Dahmen, and H. Ney. Parameter Estimation for Automatic Dose Control in Radioscopy. In Bildverarbeitung für die Medizin (BVM), pages 279-282, Leipzig, Germany, March 2002.
  
• T. M. Lehmann, S. Goudarzi, N. Linnenbrügger, D. Keysers, and B. Wein. Lokalisation und Delineation des Kollimatorfeldes in digitalen und film-basierten Röntgenbildern (in German). In Bildverarbeitung für die Medizin (BVM), pages 51-54, Leipzig, Germany, March 2002.
  
• M. O. Güld, J. Bredno, D. Keysers, M. Kohnen, C. Thies, H. Schubert, B. Wein, and T. M. Lehmann. Eine verteilte Systemarchitektur für IRMA (Image Retrieval in Medical Applications). In Bildverarbeitung für die Medizin (BVM), pages 175-178, Leipzig, Germany, March 2002.
  
• T. M. Lehmann, S. Goudarzi, N. Linnenbrügger, D. Keysers, and B. Wein. Automatic Localization and Delineation of Collimation Fields in Digital- and Film-based Radiographs. In International Society for Optical Engineering (SPIE), Proc. SPIE, pages 1215-1223, San Diego, CA, February 2002.
  
• M. O. Güld, M. Kohnen, D. Keysers, H. Schubert, B. Wein, J. Bredno, and T. M. Lehmann. Quality of DICOM header Information for Image Categorization. In Intl. Symposium on Medical Imaging, Proc. SPIE, volume 4685, pages 280-287, San Diego, CA, February 2002.
  

Page: 1

---

Acknowledgements

This work is partially funded by the DFG (Deutsche Forschungsgemeinschaft), and as part of the Quaero Programme, funded by OSEO, French State agency for innovation.

---