Patrick Irmisch
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Patrick Irmisch has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Anko Boerner
Deutsches Zentrum für Luft-und Raumfahrt (DLR) Institut für Optische Sensorsysteme Rutherfordstraße 2 12489 Berlin- Adlershof Germany
1 shared publication Anko Borner
Institute of Optical Sensor Systems, German Aerospace Center, 12489, Berlin
1 shared publication- Dirk Baumbach
German Aerospace Center
1 shared publication - Guillermo Gallego
Technical University of Berlin
1 shared publication - Ines Ernst
German Aerospace Center (DLR), Institute of Optical Sensor Systems, Berlin, Germany
1 shared publication - Maik Wischow
German Aerospace Center (DLR), Institute of Optical Sensor Systems, Berlin, Germany
1 shared publication
Latest Publications
- Open Access
- 26 April 2024
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (6)
Real‐Time Noise Source Estimation of a Camera System from an Image and Metadata
Autonomous machines must self‐maintain proper functionality to ensure the safety of humans and themselves. This pertains particularly to its cameras as predominant sensors to perceive the environment and support actions. A fundamental camera problem addressed in this study is noise. Solutions often focus on denoising images a posteriori, that is, fighting symptoms rather than root causes. However, tackling root causes requires identifying the noise sources, considering the limitations of mobile platforms. In this work, a real‐time, memory‐efficient, and reliable noise source estimator that combines data‐based and physically based models is investigated. To this end, a deep neural network that examines an image with camera metadata for major camera noise sources is built and trained. In addition, it quantifies unexpected factors that impact image noise or metadata. This study investigates seven different estimators on six datasets that include synthetic noise, real‐world noise from two camera systems, and real‐field campaigns. For these, only the model with most metadata is capable to accurately and robustly quantify all individual noise contributions. This method outperforms total image noise estimators and can be plug‐and‐play deployed. It also serves as a basis to include more advanced noise sources, or as part of an automatic countermeasure feedback loop to approach fully reliable machines.
- 26 April 2024
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (6)
Open AccessReal‐Time Noise Source Estimation of a Camera System from an Image and Metadata
Autonomous machines must self‐maintain proper functionality to ensure the safety of humans and themselves. This pertains particularly to its cameras as predominant sensors to perceive the environment and support actions. A fundamental camera problem addressed in this study is noise. Solutions often focus on denoising images a posteriori, that is, fighting symptoms rather than root causes. However, tackling root causes requires identifying the noise sources, considering the limitations of mobile platforms. In this work, a real‐time, memory‐efficient, and reliable noise source estimator that combines data‐based and physically based models is investigated. To this end, a deep neural network that examines an image with camera metadata for major camera noise sources is built and trained. In addition, it quantifies unexpected factors that impact image noise or metadata. This study investigates seven different estimators on six datasets that include synthetic noise, real‐world noise from two camera systems, and real‐field campaigns. For these, only the model with most metadata is capable to accurately and robustly quantify all individual noise contributions. This method outperforms total image noise estimators and can be plug‐and‐play deployed. It also serves as a basis to include more advanced noise sources, or as part of an automatic countermeasure feedback loop to approach fully reliable machines.
- 1 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1995-1999
Simulation Framework for a Visual-Inertial Navigation System
The ability to test methods based on simulated data is an important component in the development of intelligent localization systems at the present time. This includes the replication of realistic environments and sensors on the one hand. On the other hand, an implementation of a correct movement profile of the target mobile platform is required. This is complicated by the large variety of possible platform variants. We propose a method to transfer movement profiles that have been recorded in real world into a simulation environment. The associated modular simulation framework was designed especially, but not limiting, for the development of visual-inertial odometry of a handheld navigation system. We demonstrate the application of our pipeline through an experiment to determine the influence of motion blur on the visual odometry component.
- 1 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1995-1999
Simulation Framework for a Visual-Inertial Navigation System
The ability to test methods based on simulated data is an important component in the development of intelligent localization systems at the present time. This includes the replication of realistic environments and sensors on the one hand. On the other hand, an implementation of a correct movement profile of the target mobile platform is required. This is complicated by the large variety of possible platform variants. We propose a method to transfer movement profiles that have been recorded in real world into a simulation environment. The associated modular simulation framework was designed especially, but not limiting, for the development of visual-inertial odometry of a handheld navigation system. We demonstrate the application of our pipeline through an experiment to determine the influence of motion blur on the visual odometry component.