Knowledge-based computer aided support of the engineering process of designing, assembling, calibrating and deploying visual quality inspection software solutions.
The overall goal is to guide the engineer by providing algorithmic approaches of similar inspection problems, and to unburden the engineer from recurrent design, tuning and implementation steps
This requires a better automatic reuse of knowledge and software modules for accelerating the whole engineering process of modelling and algorithmic workflow design.
The objectives address the usability of such an engineering software framework, its productivity and the performance of the resulting software solution in terms of classification rate, robustness, need of processing power and configuration effort.
Aims / Research Topics
- Support through appropriate ontologiesfor specification, description and planning of the visual quality inspection problem.
- Support through appropriate ontologies in the selection of algorithms and workflows for image analysis, image processing and machine learning tasks.
- Configuration support and optimization of algorithmic workflows by adopting best practices of a parallelizable optimization framework.
- Support the high-performance implementation of the distilled algorithmic workflows for the targeted processing hardware systems (GPU, multi-core) by hardware-virtualization techniques.
Methods / Software / Proof of Concepts
- Research on parameter-based methods with sparse parameters for texture analysis and pattern recognition on the basis of non-standardized similarity measures such as the Discrepancy Norm.
- Application of general information theory principles.
- Improvement of the integrated workflow "Image acquisition - automated processing - user interaction", e. g. by supporting the user with additional relevant information regarding plausibility measures and recommendations based on abductive reasoning techniques.
- Proof of concepts for camera calibration in a non-standard configurations.