Zusammenfassung
Surface characteristics can significantly impact the functionality of a component. Accurately measuring and describing these characteristics therefore plays an important role in precision engineering and manufacturing fields, such as aerospace, automotive, biomedical, and semiconductor manufacturing. Optical measurement systems face a trade-off between high lateral resolution, necessary for resolving small features, and the need for a large measuring field for comprehensive results. A common solution is stitching multiple high-resolution images to effectively create a large measuring field. However, conventional registration algorithms often require extensive overlaps between individual images, lack robustness against large image variations, are susceptible to noise, and necessitate manual parameter adjustments. This work presents a comprehensive investigation and development of image registration techniques in optical metrology, aiming to improve the registration of micro-topographic image data compared to conventional methods. A key contribution is the development and validation of a novel registration approach based on convolutional neural networks, specifically a two-stage architecture named Coarse-to-Fine Image Registration (CoFiR) Net. This method enables significant improvements over conventional registration techniques in terms of accuracy, robustness against large image variations and image noise, as well as computational speed. The development and validation of the CoFiR Net are conducted using an extensive dataset of micro-topographic measurements. This dataset comprises over 70 000 measurements with two confocal laser scanning microscopes at various magnifications, on samples involving a wide range of materials, machining methods, manufacturing processes, and different surface roughnesses. This dataset offers a valuable resource for future work in areas such as defect detection, surface classification, image super-resolution, or monocular depthestimation. Additional contributions of this work include the novel use of convolutional neural networks for the registration of non-overlapping images. In conclusion, this work makes a significant contribution to surface metrology and image processing. The improvements – increased accuracy, robustness against large image variations and noise, reduced computation time, and the elimination of manual parameter adjustment – extend the application areas and utility of image registration.
Schlagworte
surface metrology image registration convolutional neural networks confocal laser scanning microscopy- Kapitel Ausklappen | EinklappenSeiten
- 1–10 1. Introduction 1–10
- 103–142 5. Results 103–142
- 147–170 Bibliography 147–170
- 171–186 Appendix 171–186
- 187–189 Curriculum Vitae 187–189
5 Treffer gefunden
- „... Convolutional Neu-ral Networks Under Conditions of Variance, Oct. 2020. arXiv: 2010.01041 [cs].[32] A. Abche, F ...” „... Con-volutional Neural Networks: A Comparison to SIFT”, no. arXiv:1405.5769,Jun. 2015. eprint: 1405.5769 (cs).[37 ...” „... , “ImageNet classificationwith deep convolutional neural networks”, Communications of the ACM,vol. 60, no. 6 ...”
- „... development and validation of a novel registration approachbased on convolutional neural networks ...” „... neural networks,confocal laser scanning microscopyviContentsList of Figures ixAcronyms xv1 Introduction ...” „... Connected Convolutional NetworksDLT direct linear transformationDOF degrees of freedomFAST features from ...”
- „... [10]–[12]. Convolutional neural networks (CNNs) have the potential to addressthese drawbacks in several ...” „... convolutional neural network(CNN)-based approaches [13]–[20]. Furthermore, typical CNNs do not cal-culate ...” „... examinationof the principles underlying neural networks (NNs) and deep learning, with aparticular emphasis on ...”
- „... weights.382.3 Machine Learning and Deep Learning2.3.3 Convolutional Neural NetworksCNNs are a category of ...” „... and optimize strategies over time in dynamicenvironments.2.3.2 Neural NetworksA NN is an ...” „... networks.Additionally, this technique expands the effective size of the receptive fields indeeper convolutional layers ...”
- „... Convolutional Networks (DenseNet)-161,• Inception-v3,• residual neural network (ResNet)-152,• VGG-19 Net with ...” „... Convolutional Networks (DenseNet) was introduced byHuang et al. from Facebook AI Research [157]. The name is ...” „... RegistrationResNetThe core idea of the residual neural network (ResNet), which was introducedby He et al. from Microsoft ...”