Deep learning for fast segmentation and critical dimension metrology & characterization enabling AR/VR design and fabrication
This work addresses the need for faster and more accurate metrology in AR/VR manufacturing, though it is incremental as it builds on existing deep learning models.
The study tackled the problem of segmenting regions of interest and extracting critical dimensions from electron microscopy images for AR/VR component fabrication by fine-tuning a pre-trained Segment Anything Model with low-rank adaptation, resulting in accurate binary image extraction and precise CD measurement.
Quantitative analysis of microscopy images is essential in the design and fabrication of components used in augmented reality/virtual reality (AR/VR) modules. However, segmenting regions of interest (ROIs) from these complex images and extracting critical dimensions (CDs) requires novel techniques, such as deep learning models which are key for actionable decisions on process, material and device optimization. In this study, we report on the fine-tuning of a pre-trained Segment Anything Model (SAM) using a diverse dataset of electron microscopy images. We employed methods such as low-rank adaptation (LoRA) to reduce training time and enhance the accuracy of ROI extraction. The model's ability to generalize to unseen images facilitates zero-shot learning and supports a CD extraction model that precisely extracts CDs from the segmented ROIs. We demonstrate the accurate extraction of binary images from cross-sectional images of surface relief gratings (SRGs) and Fresnel lenses in both single and multiclass modes. Furthermore, these binary images are used to identify transition points, aiding in the extraction of relevant CDs. The combined use of the fine-tuned segmentation model and the CD extraction model offers substantial advantages to various industrial applications by enhancing analytical capabilities, time to data and insights, and optimizing manufacturing processes.