Veröffentlichung zu Processing Pipelines für Industrielle Bildverarbeitung mittels kartesischer genetischer Programmierung erschienen / Article on Processing Pipelines for Industrial Imaging with Cartesian Genetic Programming published [25.09.23]
Die Autoren Margraf, Cui, Stein und Hähner publizieren auf der ACSOS – The authors Margraf, Cui, Stein and Hähner publish at ACSOS
ENGLISH VERSION BELOW.
Es gibt Neuigkeiten zur Forschungsarbeit am KI-Fachgebiet. Eine weitere Veröffentlichung ist zum Thema Processing Pipelines für Industrielle Bildverarbeitung mittels kartesischer genetischer Programmierung erschienen.
Hier kommen die Hard-Facts zur Veröffentlichung.
Titel:
Evolving Processing Pipelines for Industrail Imnaging with Cartesian Genetic Programming
Autoren:
Andreas Margraf, Henning Cui, Anthony Stein, Jörg Hähner
Venue:
2023 IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS)
Abstract:
The reconfiguration of machine vision systems heavily depends on the collection and availability of large datasets, rendering them inflexible and vulnerable to even minor changes in the data. This paper proposes a refinement of Miller’s Cartesian Genetic Programming methodology, aimed at generating filter pipelines for image processing tasks. The approach is based on CGP-IP, but specifically adapted for image processing in industrial monitoring applications. The suggested method allows for retraining of filter pipelines using small datasets; this concept of self-adaptivity renders high-precision machine vision more resilient to faulty machine settings or changes in the environment and provides compact programs. A dependency graph is introduced to rule out invalid pipeline solutions. Furthermore, we suggest to not only generate pipelines from scratch, but store and reapply previous solutions and re-adjust filter parameters. Our modifications are designed to increase the likelihood of early convergence and improvement in the fitness indicators. This form of self-adaptivity allows for a more resource-efficient configuration of image filter pipelines with small datasets.
Unter folgendem Link kommen Sie zur gesamten Übersicht der bisherig veröffentlichten Forschungsarbeiten des Fachgebiets: ki-agrartechnik.uni-hohenheim.de/veroeffentlichungen
--
There is news about the research work at the AI department. Another publication has appeared on the topic of Processing Pipelines for Industrial Imaging with Cartesian Genetic Programming.
Here are the hard facts about the publication.
Title:
Evolving Processing Pipelines for Industrail Imnaging with Cartesian Genetic Programming
Authors:
Andreas Margraf, Henning Cui, Anthony Stein, Jörg Hähner
Venue:
2023 IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS)
Abstract:
The reconfiguration of machine vision systems heavily depends on the collection and availability of large datasets, rendering them inflexible and vulnerable to even minor changes in the data. This paper proposes a refinement of Miller’s Cartesian Genetic Programming methodology, aimed at generating filter pipelines for image processing tasks. The approach is based on CGP-IP, but specifically adapted for image processing in industrial monitoring applications. The suggested method allows for retraining of filter pipelines using small datasets; this concept of self-adaptivity renders high-precision machine vision more resilient to faulty machine settings or changes in the environment and provides compact programs. A dependency graph is introduced to rule out invalid pipeline solutions. Furthermore, we suggest to not only generate pipelines from scratch, but store and reapply previous solutions and re-adjust filter parameters. Our modifications are designed to increase the likelihood of early convergence and improvement in the fitness indicators. This form of self-adaptivity allows for a more resource-efficient configuration of image filter pipelines with small datasets.
The following link will take you to the complete overview of previously published research papers in the field: ki-agrartechnik.uni-hohenheim.de/veroeffentlichungen
