@article{BilalPodishettiGrossmannetal.2025, author = {Bilal, M{\"u}henad and Podishetti, Ranadheer and Großmann, Daniel and Bregulla, Markus}, title = {Benchmarking CNN Architectures for Tool Classification: Evaluating CNN Performance on a Unique Dataset Generated by Novel Image Acquisition System}, volume = {13}, journal = {IEEE Access}, publisher = {IEEE}, address = {New York}, issn = {2169-3536}, doi = {https://doi.org/10.1109/ACCESS.2025.3574785}, pages = {96400 -- 96422}, year = {2025}, abstract = {In this study, we introduce the ToolSurface-144 dataset, which is presented here for the first time. It comprises four subsets - Full R, Full S, Top R, and Top S - each containing 144 tool classes captured under varying illumination conditions and fields of view. A newly developed, patented imaging approach was employed to acquire the data. It is compared with conventional diffuse ring illumination to assess its effectiveness in evaluating state-of-the-art convolutional neural networks. This enabled a more targeted investigation of the role of global shape characteristics such as silhouettes versus localized features like the tool face, cutting edges, and delicate geometrical structures under different training strategies. In this study, we evaluate six state-of-the-art convolutional neural networks—AlexNet, DenseNet161, EfficientNet-B0, ResNet152, ResNet50, and VGG16—using three training strategies: fine-tuning, freezing of pre-trained layers, and training from scratch. The results show that EfficientNet-B0 consistently achieved the highest classification accuracy in nearly all experiments and data sets. Especially using the fine-tuning training strategy, the model achieved 99\% accuracy in tool classification. ResNet50 benefited greatly from fine-tuning and freezing, achieving a significant increase in performance compared to training from scratch. In contrast, ResNet152, AlexNet, and VGG16 consistently showed poor classification performance, indicating difficulties regarding learning and generalisation. The results show that diffuse illumination and complete tool views provide the best classification conditions, while restricted image sections with homogeneous illumination negatively affect model performance. Among the evaluated training strategies, fine-tuning proved the most efficient training method for developing CNN models for tool classification.}, language = {en} } @article{BilalPodishettiKovaletal.2024, author = {Bilal, M{\"u}henad and Podishetti, Ranadheer and Koval, Leonid and Gaafar, Mahmoud A. and Großmann, Daniel and Bregulla, Markus}, title = {Automatized End Mill Wear Inspection Using a Novel Illumination Unit and Convolutional Neural Network}, volume = {12}, journal = {IEEE Access}, publisher = {IEEE}, address = {New York}, issn = {2169-3536}, doi = {https://doi.org/10.1109/ACCESS.2024.3454692}, pages = {124282 -- 124297}, year = {2024}, abstract = {Ensuring cutting tools are in optimal condition is essential for achieving peak machining performance, given their direct impact on both workpiece quality and process efficiency. However, accurately assessing wear on end mills, especially those with complex geometries, pose a significant challenge due to their reflective surfaces and varied wear patterns. Presented here is a novel method that addresses this challenge by employing a customized illumination unit in conjunction with a convolutional neural network (CNN) for end mill wear analysis. This innovative approach involves utilizing the specially designed illumination unit to capture high-quality images, enabling precise examination of material wear on helically shaped end mills. Notably, this method is tailored to illuminate reflective surfaces and represents a pioneering application in the realm of wear testing.We validate the viability of this approach by employing CNN-based models to segment wear on complex-shaped end mills coated with titanium carbonitride (TiCN) and titanium nitride (TiN). We achieved remarkable mean Intersection over Union (mIoU) results in wear detection on a test dataset: 0.99 for tool segmentation, 0.78 for abnormal wear, and 0.71 for normal wear segmentation.}, language = {en} } @inproceedings{AkmalAsifKovaletal.2025, author = {Akmal, Muhammad Uzair and Asif, Saara and Koval, Leonid and Mathias, Selvine George and Knollmeyer, Simon and Großmann, Daniel}, title = {Layered Data-Centric AI to Streamline Data Quality Practices for Enhanced Automation}, booktitle = {Artificial Intelligence: Methodology, Systems, and Applications: 19th International Conference, AIMSA 2024, Varna, Bulgaria, September 18-20, 2024, Proceedings}, editor = {Koprinkova-Hristova, Petia and Kasabov, Nikola}, publisher = {Springer}, address = {Cham}, isbn = {978-3-031-81542-3}, doi = {https://doi.org/10.1007/978-3-031-81542-3_11}, pages = {128 -- 142}, year = {2025}, language = {en} } @inproceedings{MathiasAsifAkmaletal.2025, author = {Mathias, Selvine George and Asif, Saara and Akmal, Muhammad Uzair and Knollmeyer, Simon and Koval, Leonid and Großmann, Daniel}, title = {Industrial Image Grouping Through Pre-Trained CNN Encoder-Based Feature Extraction and Sub-Clustering}, booktitle = {Proceedings of the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025) - Volume 2}, editor = {Rocha, Ana Paula and Steels, Luc and van den Herik, Jaap}, publisher = {SciTePress}, address = {Set{\´u}bal}, isbn = {978-989-758-737-5}, doi = {https://doi.org/10.5220/0013189000003890}, pages = {496 -- 506}, year = {2025}, abstract = {A common challenge faced by many industries today is the classification of unlabeled image data from production processes into meaningful groups or patterns for better documentation and analysis. This paper presents a sequential approach for leveraging industrial image data to identify patterns in products or processes for plant floor operators. The dataset used is sourced from steel production, and the model architecture integrates feature reduction through convolutional neural networks (CNNs) like VGG, EfficientNet, and ResNet, followed by clustering algorithms to assign appropriate labels to the observed data. The model's selection criteria combine clustering metrics, including entropy minimization and silhouette score maximization. Once primary clusters are identified, sub-clustering is performed using near-labels, which are pre-assigned to images with initial distinctions. A novel metric, C-Score, is introduced to assess cluster convergence and grouping accuracy. Experimental re sults demonstrate that this method can address challenges in detecting variations across images, improving pattern recognition and classification.}, language = {en} } @article{MathiasAkmalAsifetal.2024, author = {Mathias, Selvine George and Akmal, Muhammad Uzair and Asif, Saara and Koval, Leonid and Knollmeyer, Simon and Großmann, Daniel}, title = {Pattern Identifications in Transformed Acoustic Signals Using Classification Models}, volume = {2024}, journal = {Procedia CIRP}, number = {130}, publisher = {Elsevier}, address = {Amsterdam}, issn = {2212-8271}, doi = {https://doi.org/10.1016/j.procir.2024.10.061}, pages = {93 -- 99}, year = {2024}, abstract = {Pattern identifications in signals is necessary to discern variations from approved normal values in different scenarios. With machine learning algorithms, it is possible to use hybrid methods of pattern identifications such as feature extractions followed by classifications and/or clustering. This paper presents a pattern identification approach of acoustic signals using their transformations as inputs to classification algorithms. The analysis is carried out on two transformed versions of acoustic emission (AE) hits such as log transformations of peak hits and binary sequencing based on threshold crossing. A comparative analysis using custom data loss metrics is made to determine which inputs provide the best information in predictive methods for identifying commonly occurring patterns while acknowledging significant data loss. The methodology is conducted on transformed versions of a public dataset and the results show that patterns can be discerned to above 90\% accuracy with the transformed datasets. The experimental results yield that actual source signals need not be utilized depending on which transformations suit the practical application.}, language = {en} } @inproceedings{KnollmeyerAkmalKovaletal.2024, author = {Knollmeyer, Simon and Akmal, Muhammad Uzair and Koval, Leonid and Asif, Saara and Mathias, Selvine George and Großmann, Daniel}, title = {Document Knowledge Graph to Enhance Question Answering with Retrieval Augmented Generation}, booktitle = {2024 IEEE 29th International Conference on Emerging Technologies and Factory Automation (ETFA)}, editor = {Facchinetti, Tullio and Cenedese, Angelo and Lo Bello, Lucia and Vitturi, Stefano and Sauter, Thilo and Tramarin, Federico}, publisher = {IEEE}, address = {Piscataway}, isbn = {979-8-3503-6123-0}, doi = {https://doi.org/10.1109/ETFA61755.2024.10711054}, year = {2024}, language = {en} } @inproceedings{AsifAkmalKovaletal.2024, author = {Asif, Saara and Akmal, Muhammad Uzair and Koval, Leonid and Knollmeyer, Simon and Mathias, Selvine George and Großmann, Daniel}, title = {Supervised Anomaly Detection for Production Line Images using Data Augmentation and Convolutional Neural Network}, booktitle = {2024 IEEE 29th International Conference on Emerging Technologies and Factory Automation (ETFA)}, editor = {Facchinetti, Tullio and Cenedese, Angelo and Lo Bello, Lucia and Vitturi, Stefano and Sauter, Thilo and Tramarin, Federico}, publisher = {IEEE}, address = {Piscataway}, isbn = {979-8-3503-6123-0}, doi = {https://doi.org/10.1109/ETFA61755.2024.10710718}, year = {2024}, language = {en} } @inproceedings{FleischmannFriedlGrossmannetal.2021, author = {Fleischmann, Albert and Friedl, Anton and Großmann, Daniel and Schmidt, Werner}, title = {Modeling and implementing of industrie 4.0 scenarios}, booktitle = {Modelling to Program}, publisher = {Springer}, address = {Cham}, isbn = {978-3-030-72695-9}, issn = {1865-0929}, doi = {https://doi.org/10.1007/978-3-030-72696-6_4}, pages = {90 -- 112}, year = {2021}, language = {en} } @inproceedings{KovalAkmalAsifetal.2025, author = {Koval, Leonid and Akmal, Muhammad Uzair and Asif, Saara and Mathias, Selvine George and Knollmeyer, Simon and Großmann, Daniel}, title = {Optimizing AI-Driven Production in Industry 4.0: A Morphological Box and Taxonomy Approach}, booktitle = {2025 International Conference on Computer Technology Applications (ICCTA)}, publisher = {IEEE}, address = {Piscataway}, isbn = {979-8-3315-1265-1}, doi = {https://doi.org/10.1109/ICCTA65425.2025.11166106}, pages = {272 -- 278}, year = {2025}, language = {en} } @inproceedings{KovalAkmalAsifetal.2024, author = {Koval, Leonid and Akmal, Muhammad Uzair and Asif, Saara and Mathias, Selvine George and Knollmeyer, Simon and Großmann, Daniel}, title = {Addressing the complexity of AI Integration in Manufacturing: A Morphological Analysis}, booktitle = {2024 IEEE 29th International Conference on Emerging Technologies and Factory Automation (ETFA)}, editor = {Facchinetti, Tullio and Cenedese, Angelo and Lo Bello, Lucia and Vitturi, Stefano and Sauter, Thilo and Tramarin, Federico}, publisher = {IEEE}, address = {Piscataway}, isbn = {979-8-3503-6123-0}, doi = {https://doi.org/10.1109/ETFA61755.2024.10711011}, year = {2024}, language = {en} } @inproceedings{AsifAkmalKovaletal.2024, author = {Asif, Saara and Akmal, Muhammad Uzair and Koval, Leonid and Mathias, Selvine George and Knollmeyer, Simon and Großmann, Daniel}, title = {A Conceptual Framework for Addressing Class Imbalance in Image Data: Challenges and Strategies}, booktitle = {2024 IEEE 12th International Conference on Intelligent Systems (IS): Proceedings}, editor = {Sgurev, Vassil and Jotsov, Vladimir and Piuri, Vincenzo and Doukovska, Luybka and Yoshinov, Radoslav}, publisher = {IEEE}, address = {Piscataway}, isbn = {979-8-3503-5098-2}, doi = {https://doi.org/10.1109/IS61756.2024.10705251}, year = {2024}, language = {en} } @inproceedings{KovalWaechterErdoganetal.2025, author = {Koval, Leonid and W{\"a}chter, Sonja and Erdogan, H{\"u}seyin and Großmann, Daniel}, title = {Ontology-Driven Modeling and Integration of Production Processes in Advanced Driver-Assistance Systems within the Gaia-X Ecosystem}, booktitle = {2025 11th International Conference on Computer Technology Applications (ICCTA 2025)}, publisher = {IEEE}, address = {Piscataway}, isbn = {979-8-3315-1265-1}, doi = {https://doi.org/10.1109/ICCTA65425.2025.11166225}, pages = {211 -- 216}, year = {2025}, language = {en} } @inproceedings{NowackiMuellerGrossmannetal.2023, author = {Nowacki, Natalie Samanta and Mueller, Ralph and Großmann, Daniel and Lueder, Arndt}, title = {Approach for identifying data usage information objects for the later implementation in production environments}, booktitle = {2023 The 10th International Conference on Industrial Engineering and Applications (Europe), ICIEA-EU 2023}, subtitle = {Enabling a logic for the information modelling based on use cases using a conversation concept}, publisher = {ACM}, address = {New York}, isbn = {978-1-4503-9852-7}, doi = {https://doi.org/10.1145/3587889.3587915}, pages = {169 -- 175}, year = {2023}, language = {en} } @article{MathiasGrossmann2022, author = {Mathias, Selvine George and Großmann, Daniel}, title = {Use Cases of Data Reduction to Time Series Data in Sensor Monitoring}, volume = {16}, journal = {Journal of Ubiquitous Systems and Pervasive Networks}, number = {2}, publisher = {IASKS}, address = {[s. l.]}, issn = {1923-7332}, doi = {https://doi.org/10.5383/JUSPN.16.02.005}, pages = {87 -- 92}, year = {2022}, language = {en} } @article{SchmiedMathiasGrossmannetal.2021, author = {Schmied, Sebastian and Mathias, Selvine George and Großmann, Daniel and Mueller, Ralph and Jumar, Ulrich}, title = {Information modelling with focus on existing manufacturing systems}, volume = {2021}, journal = {Annual Reviews in Control}, number = {51}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1367-5788}, doi = {https://doi.org/10.1016/j.arcontrol.2021.04.010}, pages = {392 -- 400}, year = {2021}, language = {en} }