Optical Instrument Science, Technology, and Applications III

Rudolf Kingslake is widely regarded as one of the founders of modern optical design. When educating his students at The Institute of Optics, Professor Kingslake championed the importance of lens design fundamentals as a complement to computer-aided design. At that time, ray tracing speed was a major bottleneck in the lens design process. Now that lens designers can trace rays in fractions of a second and have access to powerful computational tools like global optimization and AI are these same fundamentals needed? Should we keep teaching them? One of Kingslake’s biggest fears was that we would forget “our laboriously acquired knowledge of geometrical optics and substitute for it the mathematical problem of optimizing a merit function”. There is no question that computers have done wonders for lens design and have enabled far more advanced designs than thought possible. The issue at hand is if mastery of both lens design fundamentals and computer software is required for success. Unfortunately, the current educational landscape places much more emphasis on the latter than the former, and many of the fundamentals impressed by Kingslake have been lost. However, three boxes of index cards belonging to Rudolf Kingslake were recently uncovered. Included in the collection are 171 lens design exam problems which present a fascinating perspective on lens design as it was taught in the pre-computer age. In this talk we’ll take a closer look at several of these forgotten problems and discuss how their solutions are still relevant for modern lens design today.

In 2017, the European Southern Observatory (ESO) awarded a contract for the Polishing, integration and final figuring of the Segment Assemblies of the primary mirror (M1) for the Extremely Large Telescope (ELT) to Safran Reosc. Since then, the design and commissioning of a production unit dedicated to ELT M1 has been accomplished and the plant has been producing many mirrors since spring 2022. We will introduce the smart factory, its processes and their automation that enabled reaching the current throughput of one mirror per day. We will then present the status of the project, some lessons learned and highlight the successes that have been achieved so far.

The development of a useful hybrid model for spectral analysis in climate change is the aim of this study. The subject of research is the dynamics tracked by the hybrid model for spectral analysis of unregulated and regulated landfills and for this purpose a database of several identical climatic seasons has been created and processed in order to verify and validate the research, which is based on satellite data , data from field surveys and unmanned aerial vehicles. The study covers examples of the different NUTS 2 planning areas (under the Regional Development and Improvement Act). Generated data of high value according to the European Commission over a period of at least five years. Regulated landfills are of national importance and selected events from the territory of Bulgaria have been investigated and monitored through a complex approach based on satellite data, UAS and ground-based innovative spectrometry equipment, NDGI, TCT, Sentinel 3 SLTRS.

Microscopy is essential in academic and medical research, driving the quest for cost-effective solutions. Inspired by modified 3D printers, these innovations enable the scanning of 2D and 3D samples, leveraging affordability and three-dimensional movement. In the realm of multimodal imaging, LED panels and LCD screens offer adaptable lighting options, balancing cost, ergonomics, and image quality remains a formidable challenge. Introducing OpenMIC, a groundbreaking solution seamlessly integrating multiple scanning features and multimodal lighting. Transformed from a 3D printer using a Raspberry Pi 4, a 64x64 LED array, and an optical module, it offers four lighting modalities, micrometric autofocus, focus stacking, image stitching, and automated scanning of 12 histological slides. OpenMIC is marked by its scalability, professional imaging quality, and a manufacturing cost under 4,000 euros.

Spectrum data obtained from hyperspectral optical systems were analyzed with a CNN-based deep learning model to detect and identify maritime small objects. The hyperspectral data set for learning was extracted from more than 60 aerial observation images, and classification accuracy was derived by applying a total of 7 CNN models. Among the models used, Inception_v3 was the best at 94.9%, and this result showed more than 10% improvement in accuracy over previous studies conducted with multi-layer perceptron (MLP). If further research breaks down classification items and increases the size of datasets, we expect that the technology will become increasingly utilized in the field of maritime search and surveillance.

This paper presents an improved WFS design that reaches both previous goals. The device is named Crossed-sine phase sensor (CSPS) and is based on a fully transparent Gradient phase filter (GPF) whose virtual image is located forward or backward the exit pupil of the tested optical system. After briefly summarizing the theoretical principle of the CSPS, we describe four different optical layouts meeting its requirements and evaluate them with the help of Zemax modeling. We finally select the best optical design and discuss the achieved optical performancesources

This poster presents a comprehensive overview of the end-to-end optical design of GRANCAIN. It explores the selection criteria for diverse commercial elements, thermal analysis utilizing Ansys Zemax OpticStudio and the acceptance tests performed at the IAC. Also encompasses tolerance analysis using Ansys Zemax OpticStudio, along with the establishment of the error budget. Furthermore, the text provides a detailed account of the alignment process, achieved through the mechanical positioning of each optical element with a laser tracker and the confirmation of positions under cryogenic conditions is conducted using an alignment telescope. Lastly, the article discusses the optical acceptance plans for the instrument before its integration into GTCAO.

This paper explores the integration of Whispering Gallery Mode (WGM) technology in parkour-style humanoid robots, specifically focusing on flywheel synchronization for smooth landings. By monitoring the WGM spectrum, the rotational speed and phase shift of the flywheel are tracked to synchronize braking during the robot's landing. This approach aims to ensure controlled touchdowns, reducing impact forces and enhancing stability. The study highlights the potential of WGM-based sensing and control mechanisms to optimize agility and precision in humanoid robots performing dynamic maneuvers.

The ASTRABAX project ("Aschaffenburg Stratospheric Balloon Experiment"), being funded for the years 2024 to 2026, is designed as a multimodal platform for the investigation of radiation exposures at high altitude. The UV-C spectral region is of special interest. Spectral measurements observe this region of interest using miniature UV-VIS spectrometers. The platform also contains a radiation dosimetry, a power source for on-board electronics, and common shielding setups for multiple spectral combinations. Human cells are exposed simultaneously to radiation of different compositions as particle, X-ray and UV radiation. After the flight, possible changes in the spatial chromatin organization are examined. Material samples intended for the development of satellite components are irradiated also. Investigations under such conditions are realistic and crucial for high altitude flights in the atmosphere, for space flights as well as for comparable exposures on other objects of the solar system, and even for exoplanet habitats to some extent.

This project aims to formulate, design , build and test a versatile, high-efficiency, low-resolution spectrograph to function as the G-CLEF (GMT-Consortium Large Earth Finder) exposure meter. G-CLEF, the first-generation Giant Magellan Telescope's (GMT) instrument, is a state-of-the-art, high-resolution, echelle spectrograph for the GMT, expected to be completed for the telescope's first light. The exposure meter plays a vital role for adjusting barycentric corrections of Doppler radial velocity (RV) by accounting for Earth's chromatic atmospheric influences. Its significance becomes pronounced in Extreme Precision RV (EPRV) measurements, where the atmosphere's wavelength dependency contributes to errors at the scale of tens of centimeters per second, the same level of precision required for detecting Earth-analog planets orbiting stars similar to the Sun, aligning with one of the primary scientific objectives of G-CLEF. This paper explores the scientific motivation in detail, describes the designs trade-off analysis and the performance simulations aiming to achieve 1cm/s precision on EPRV measurements and outlines the resulting principal parameters derived from these analyses.

In current camera sensor-objective alignment approaches for an automotive industry, slow and iterative scan methods are widely used with a cycle time not fully compatible with contemporary production lines. Our proposed alignment method based on a combination of classic centroid based and AI based calculation facilitates a single-shot operation. This immediately leads to massive production line simplification and cost reduction, so important for the automotive industry.

Asphere and freeform metrology forms the basis of precision optics fabrication. Stitching or scanning methods provide the necessary flexibility, but require measurement times of several minutes. Using parallel information channels of light (wavelength, polarization, phase) in combination with the model-based tilted wave interferometry approach boosts the measurement possibilities. In this proceeding we show how these information channels can be used to enhance the measurement capabilities regarding reconstruction quality and show how the complete shape information of strong aspheres can be recorded by TWISS (tilted wave interferometry single-shot) within milliseconds.

In this study we have designed, assembled, and characterized a wavefront sensor that works with defocused intensity images and the wavefront phase imaging (WFPI) algorithm. This approach allows for the potential utilization of the entire sensor surface, enabling high-resolution operation. This sensor, equipped with an electrically tuneable lens (ETL), performs focus movements of more than 60 Hz, enough for real time applications. We have developed numerical tools, as a practical software environment, with a graphical user interface (GUI), to make the camera a versatile instrument easily adaptable to different experimental setups without drastic changes in the optical configuration. These tools allow to analyse the wavefront in real time to extract the desired metrics and results.

The refractive index structure constant and the microscale of turbulence contribute as crucial indicators of turbulence strength in a medium, such as the atmosphere. These parameters exert a substantial influence on measurements performed with Laser Doppler Vibrometry in an open atmosphere. In this publication, we measure these turbulent parameters with high precision through the application of the Rytov's method of smooth perturbations. This approach involves analyzing three distinct theoretical scenarios based on the separation distance between the parallel laser beams. Using a multipoint Laser Doppler Vibrometer (MPV), we estimated turbulence intensity parameters by applying the smooth perturbation method to the proposed solutions of the stochastic wave equation.

In this work, we exploit HHG in a noble gas to merge the azimuthally twisted wavefront of a vortex beam and the spatially varying polarization of a vector beam, yielding EUV vector-vortex beams (VVB) that are tailored simultaneously in their SAM and OAM. Employing a high-resolution EUV Hartmann wavefront sensor (EUV HASO, Imagine Optic), we perform the complete spatial intensity and wavefront characterization of the vertical polarization component of the 25th harmonic beam centered at a wavelength of 32.6 nm. By driving the HHG using IR VVB, we show that HHG enables the production of EUV VVB exhibiting radial, azimuthal, or even intermediate polarization distribution. Furthermore, the wavefront characterization allows for the unambiguous confirmation of the topological charge and OAM helicity of the upconverted harmonic VVB. Notably, our work reveals that HHG provides a means for the synchronous and controlled manipulation of SAM and OAM. The production of ultrafast EUV VVB with high OAM and adjustable polarization distributions opens up promising prospects for their applications at nanometric spatial and sub-femtosecond temporal resolutions using a table-top harmonic source.

The upcoming IEEE standard P4001 outlines a set of parameters for characterizing the performance of hyperspectral cameras along with recommended measurement procedures. This study concentrates on validating a simplified approach to measure the across-track spatial resolution of a pushbroom hyperspectral camera with a sensor sampling factor of two or higher, in comparison to the scanning-based approach described in the standard. The findings indicate that the snapshot-based method produces values for the width of the sampling line spread function in the across-track direction that closely match those obtained through scanning with sub-pixel steps for hyperspectral cameras with a sensor sampling factor of two or higher.

Conventional agriculture relies heavily on herbicides for weed control. Smart farming, particularly through the use of mechanical weed control systems, has the potential to reduce the herbicide usage and the associated negative impact on our environment. The growing accessibility of multispectral cameras in recent times poses the question if their added expenses justify the potential advantages they offer. In this study we compare the weed and crop detection performance between RGB and multispectral VIS-NIR imaging data. Therefore, we created and annotated a multispectral instance segmentation dataset for sugar beet crop and weed detection. We trained Mask-RCNN models on the RGB images and on images composed of different vegetation indices calculated from the multispectral data. The outcomes are thoroughly analysed and compared across various scenarios. Our findings indicate that the use of vegetation indices can significantly improve the weed detection performance in many situations.

In the realm of high-throughput manufacturing, optical metrology tools are indispensable for inspecting surface defects and ensuring quality control. This paper introduces a fast, beam-scanning version of the optical metrology and inspection technique: Coherent Fourier Scatterometry (CFS). Unlike previous designs employing piezo scanners for sample raster scanning while maintaining a constant beam, this paper presents a novel approach utilizing a galvo mirror for rapid beam scanning along the fast axis and a slower piezo stage along the slow axis. The article provides a comprehensive overview of the optical system design, considerations for alignment, a discourse on optical aberrations affecting the measurement signal, and the calibration of the system using known structures

Production and qualification of electronic cameras currently requires the use of optical target projectors. The quality of the target projectors used impacts the quality of the final products. We designed and constructed an apparatus for evaluating target projector parameters, such as the projection distance, target backlighting, projector FoV and projection quality. This device allows us to qualify optical target projectors and to perform long term and under stress tests.

We employ aligned two-photon lithography (A2PL®), to push high precision alignment tasks from the photonic packaging step towards the fabrication process in a one step process. Combined with Two-Photon Grayscale Lithography (2GL®), this approach enables direct fabrication of micro-optical elements onto devices, enhancing functionality with highest surface quality and shape fidelity at high throughput. We demonstrate automated 3D alignment using customer-ready detection algorithms, to fabricate micro-optical elements attached to various topographies and material platforms with exceptional accuracy below 100 nm. We showcase micro-optical elements aligned to fiber tips, photonic edge couplers, and photonic grating couplers to demonstrate the validity of A2PL for improved coupling losses and beam quality.

AI-based methods are increasingly crucial in industrial optical systems for tasks like identification and classification. They excel in complex classification tasks, but acquiring suitable training data is challenging. This study uses an AI-based optical scanner designed for cast part identification in the casting industry to investigate the influence of synthetic training data on classification performance. The scanner reads beveled cylindrical pin orientations to identify specific cast parts. To create synthetic training datasets, Physically Based Rendering (PBR) is used. This approach allows automatic labeling of the rendered images and a systematic investigation of the importance of the rendering parameters on the recognition rate for real image data. Promising classification results are achieved with synthetic training datasets, only.

This study delves into integrating Whispering Gallery Mode (WGM) microresonators in jumping humanoid robots to improve landing impact detection and control strategies. By situating WGM sensors near landing points, the research investigates WGM frequency shifts caused by impact forces during landing. Adaptive control algorithms utilize this data to dynamically adjust leg stiffness and joint angles for enhanced stability on varied terrains. Additionally, the research optimizes rotation rates and release timings based on WGM feedback to achieve desired jump heights while conserving energy. This integration demonstrates potential for improving agility and efficiency in humanoid robot locomotion.

Compact-sized microscopy is revolutionizing object observation. Miniature microscopes enable continuous monitoring by attaching one to each specimen, allowing for parallel analysis. This innovation is invaluable for applications like drug discovery with organ-on-a-chip devices, where numerous drug/sample pairs need assessment. Previously, compact microscopes were limited to brightfield techniques, excluding powerful tools like fluorescent microscopy. In this breakthrough, we introduce a miniature microscope with integrated fluorescence measurement. It combines a 10 μm-diameter single-photon avalanche diode and a 640 x 480 InGaN/GaN microdisplay. This advancement yields life-science-suitable images and significantly faster operation compared to existing compact brightfield microscopes.

Microendoscopy systems are extensively researched for high-resolution imaging and disease diagnosis. While commercial endoscopes employ fiber bundles, their cost and resolution constraints pose challenges. Fiber scanning-based systems offer a promising alternative, but face difficulties in aligning free space optics and photomultiplier tubes for fluorescence detection. Our proposed solution utilizes a compact and cost-effective design incorporating double-clad fiber of smaller inner clad, an Avalanche Photodetector, and a bandpass filter for fluorescence signal detection. We experimentally validated the system's performance using a piezoelectric tube-actuated fiber-scanning setup with a cylindrical GRIN lens, demonstrating effective reflectance and fluorescence imaging at visible wavelengths.

Recent technological advances have paved the way for real-time, near-video volumetric optical coherence tomography (OCT) signal acquisition and reconstruction, termed four-dimensional microscope-integrated OCT (4D-miOCT). We present a state-of-the-art 4D-miOCT engine capable of megahertz OCT data acquisition and display, and its potential clinical applications. A switchable light source at the heart of the engine enables imaging depths that allow visualizing the entire length of an eye in a single shot with micrometer resolution for capturing scans and intraoperative biometry measurements. Dynamic visualization of a surgical scene, especially the posterior segment, is possible by continuously scanning the same scene at faster A-scan rates. In addition, postoperative changes can be analyzed by exporting OCT volume data. To bridge the assessment of the applicability of 4D OCT for human surgery, we performed measurements in an experimental eye model and in ex vivo porcine eyes.

Remote sensing is a very promising technology and technique, which has become essential for various applications like agriculture, autonomous vehicles etc. Our primary focus is active remote sensing for which Lidar proved itself a very promising technology. To get a large field of view, high resolution and small spot size at the same time is always challenging. The focus is to develop a Lidar system using state of the art simulation tools to fulfil the demands of remote sensing, there is a need of development and optimization of Lidar systems that should be compact, easy and implementable for specified applications.

Numerous biological processes and microstructures remain imperceptible to the unaided eye. With the help of a microscope, students can explore the intricate details of microstructures, enriching their comprehension and knowledge retention. Biology courses encourage hands-on microscope experiments, facilitating a deeper understanding of scientific concepts. Augmented reality (AR) has witnessed remarkable advancements, ushering in heightened productivity and interactivity. AR emerges as a valuable tool for bolstering students' practical competencies. This study introduces an educational system utilizing AR technology to enhance biological microscope-based learning. It enables users to complete experiments, including identifying blood components, under our designed compound microscope. AR augments real-time information, enhancing traditional compound microscopes' performance and affordability. Our portable microscope stands as a cost-effective, user-friendly, and efficient alternative to the traditional model.