Michał Kopczyński
Research Subjects
Michał Kopczyński has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Jacek Niklewski
DAC.Digital S.A., Al. Grunwaldzka 472, 80-309 Gdańsk, Poland
3 shared publications - Krzysztof Wołk
DAC.Next Sp. z o.o., Al. Grunwaldzka 472, 80-309 Gdańsk, Poland
3 shared publications - Marek Sylwester Tatara
DAC.Digital S.A., Al. Grunwaldzka 472, 80-309 Gdańsk, Poland
3 shared publications - Oleg Żero
DAC.Digital S. A., Al. Grunwaldzka 472, 80-309 Gdańsk, Poland
1 shared publication - Piotr Kopa-Ostrowski
DAC.Digital S.A., 80-309 Gdansk, Poland
1 shared publication
Latest Publications
- Open Access
- 31 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (15) , 3370
A Human-Centric MLOps Blueprint for Visual Quality Control as a Service in an Open Access Platform
Visual quality control increasingly relies on machine learning (ML), but reliable production adoption requires a governed data-model-service lifecycle rather than an isolated detector. This article presents an implementation-grounded Open Access Platform (OAP) blueprint and audit protocol for visual quality control as a service. The contribution is not a new detector backbone. It is a reproducible governance pattern that connects ontology-based data validation, Neo4j knowledge-graph traceability, Eclipse Arrowhead service discovery, human-in-the-loop annotation, dataset and model registries, operational quality gates, and controlled model adaptation. The platform context is supported by implemented OAP tools for battery carbon-footprint analysis and production planning, while the visual inspection modules are reported at the notebook level. The benchmark protocol uses leakage-safe splits and OK samples with empty label files, separates the MVTec AD augmentation corpus from the unlearning stress-test corpus, and requires FP-on-OK gating with a retrained post-removal reference. The resulting architecture supports semantic traceability, controlled promotion and rollback, human accountability and Industry 5.0-aligned industrial AI governance without overstating the completed evidence. Compared with existing MLOps architectures, OAPs add inspection-specific semantic enforcement, knowledge-graph lineage, and governed adaptation.
- 31 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (15) , 3370
Open AccessA Human-Centric MLOps Blueprint for Visual Quality Control as a Service in an Open Access Platform
Visual quality control increasingly relies on machine learning (ML), but reliable production adoption requires a governed data-model-service lifecycle rather than an isolated detector. This article presents an implementation-grounded Open Access Platform (OAP) blueprint and audit protocol for visual quality control as a service. The contribution is not a new detector backbone. It is a reproducible governance pattern that connects ontology-based data validation, Neo4j knowledge-graph traceability, Eclipse Arrowhead service discovery, human-in-the-loop annotation, dataset and model registries, operational quality gates, and controlled model adaptation. The platform context is supported by implemented OAP tools for battery carbon-footprint analysis and production planning, while the visual inspection modules are reported at the notebook level. The benchmark protocol uses leakage-safe splits and OK samples with empty label files, separates the MVTec AD augmentation corpus from the unlearning stress-test corpus, and requires FP-on-OK gating with a retrained post-removal reference. The resulting architecture supports semantic traceability, controlled promotion and rollback, human accountability and Industry 5.0-aligned industrial AI governance without overstating the completed evidence. Compared with existing MLOps architectures, OAPs add inspection-specific semantic enforcement, knowledge-graph lineage, and governed adaptation.
- Open Access
- 22 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (14) , 3237
Multimodal Wearable Biosensing and Edge AI for Personalized Health: A Comprehensive Review
Wearable biosensing is moving beyond single-signal activity tracking toward multimodal, AI-assisted health monitoring that combines biophysical streams with biochemical information from sweat, interstitial fluid, tears, and other accessible biofluids. Recent work has accelerated progress in flexible optical materials, programmable DNA-based sensing architectures, biosafety-aware sweat patches, and edge AI pipelines capable of denoising, calibration, personalization, and low-latency inference. This review synthesizes current advances across general biosensor platforms, vital-sign monitoring, biochemical sweat sensing, motion and biomechanics sensing, and edge AI/data analytics. Particular attention is given to the translational bottlenecks that now dominate the field, including motion artifacts, sensor drift, biofouling, subject-to-subject variability, limited sweat-to-blood equivalence, insufficient external validation, and uneven regulatory readiness. The central argument of this updated review is that the next phase of progress will not be driven by sensitivity alone but by robust multimodal fusion, clinically anchored validation, interoperable data pipelines, and energy-efficient on-device intelligence. By linking materials, electronics, algorithms, and deployment constraints, the review identifies the wearable biosensing strategies most likely to progress from promising laboratory demonstrations to reliable personalized-health tools.
- 22 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (14) , 3237
Open AccessMultimodal Wearable Biosensing and Edge AI for Personalized Health: A Comprehensive Review
Wearable biosensing is moving beyond single-signal activity tracking toward multimodal, AI-assisted health monitoring that combines biophysical streams with biochemical information from sweat, interstitial fluid, tears, and other accessible biofluids. Recent work has accelerated progress in flexible optical materials, programmable DNA-based sensing architectures, biosafety-aware sweat patches, and edge AI pipelines capable of denoising, calibration, personalization, and low-latency inference. This review synthesizes current advances across general biosensor platforms, vital-sign monitoring, biochemical sweat sensing, motion and biomechanics sensing, and edge AI/data analytics. Particular attention is given to the translational bottlenecks that now dominate the field, including motion artifacts, sensor drift, biofouling, subject-to-subject variability, limited sweat-to-blood equivalence, insufficient external validation, and uneven regulatory readiness. The central argument of this updated review is that the next phase of progress will not be driven by sensitivity alone but by robust multimodal fusion, clinically anchored validation, interoperable data pipelines, and energy-efficient on-device intelligence. By linking materials, electronics, algorithms, and deployment constraints, the review identifies the wearable biosensing strategies most likely to progress from promising laboratory demonstrations to reliable personalized-health tools.
- Open Access
Forestry Segmentation Using Depth Information: A Method for Cost Saving, Preservation, and Accuracy
Forests are critical ecosystems, supporting biodiversity, economic resources, and climate regulation. The traditional techniques applied in forestry segmentation based on RGB photos struggle in challenging circumstances, such as fluctuating lighting, occlusions, and densely overlapping structures, which results in imprecise tree detection and categorization. Despite their effectiveness, semantic segmentation models have trouble recognizing trees apart from background objects in cluttered surroundings. In order to overcome these restrictions, this study advances forestry management by integrating depth information into the YOLOv8 segmentation model using the FinnForest dataset. Results show significant improvements in detection accuracy, particularly for spruce trees, where mAP50 increased from 0.778 to 0.848 and mAP50-95 from 0.472 to 0.523. These findings demonstrate the potential of depth-enhanced models to overcome the limitations of traditional RGB-based segmentation, particularly in complex forest environments with overlapping structures. Depth-enhanced semantic segmentation enables precise mapping of tree species, health, and spatial arrangements, critical for habitat analysis, wildfire risk assessment, and sustainable resource management. By addressing the challenges of size, distance, and lighting variations, this approach supports accurate forest monitoring, improved resource conservation, and automated decision-making in forestry. This research highlights the transformative potential of depth integration in segmentation models, laying a foundation for broader applications in forestry and environmental conservation. Future studies could expand dataset diversity, explore alternative depth technologies like LiDAR, and benchmark against other architectures to enhance performance and adaptability further.
Open AccessForestry Segmentation Using Depth Information: A Method for Cost Saving, Preservation, and Accuracy
Forests are critical ecosystems, supporting biodiversity, economic resources, and climate regulation. The traditional techniques applied in forestry segmentation based on RGB photos struggle in challenging circumstances, such as fluctuating lighting, occlusions, and densely overlapping structures, which results in imprecise tree detection and categorization. Despite their effectiveness, semantic segmentation models have trouble recognizing trees apart from background objects in cluttered surroundings. In order to overcome these restrictions, this study advances forestry management by integrating depth information into the YOLOv8 segmentation model using the FinnForest dataset. Results show significant improvements in detection accuracy, particularly for spruce trees, where mAP50 increased from 0.778 to 0.848 and mAP50-95 from 0.472 to 0.523. These findings demonstrate the potential of depth-enhanced models to overcome the limitations of traditional RGB-based segmentation, particularly in complex forest environments with overlapping structures. Depth-enhanced semantic segmentation enables precise mapping of tree species, health, and spatial arrangements, critical for habitat analysis, wildfire risk assessment, and sustainable resource management. By addressing the challenges of size, distance, and lighting variations, this approach supports accurate forest monitoring, improved resource conservation, and automated decision-making in forestry. This research highlights the transformative potential of depth integration in segmentation models, laying a foundation for broader applications in forestry and environmental conservation. Future studies could expand dataset diversity, explore alternative depth technologies like LiDAR, and benchmark against other architectures to enhance performance and adaptability further.