CeramOptec (Germany)

The tricky part in deploying AI models to production is often training, with two important prerequisites: One, training data must be representative of the data that the algorithm will see in later use. And two, training data must be properly labeled manually. In algorithms for automated optical inspection, there is a further problem: What if there are only a few examples of specific defect types? We tackled these problems with different strategies when developing our ARGOS system for scratch-dig inspection. We will present real-world examples of how AI algorithms can be used for defect detection and classification without large training databases.

Manual visual inspection is the standard method to evaluate cleanliness, scratches, and digs on optical components according to ISO 10110-7 or MIL-PRF-13830B. However, the limited optical resolution of the human eye turns the inspection of critical samples like high power optics or sensor cover glasses into a time-consuming and tiring task with inherently subjective results. Machine vision systems provide an objective alternative with reliable and reproducible results and additional advantages: Documentation in the form of inspection reports can be generated at no extra cost and data can be collected to analyze typical problems in the manufacturing process. Automated inspection systems need high resolution and well-designed illumination for the inspected part to ensure visibility of smallest imperfections with sizes as small as few micrometers. At the same time, short duty cycles of the inspection system are a key user requirement. Manual-visual is performed in dark field conditions. In automated inspection systems, a dark field optical setup is also advantageous, giving the best imperfection contrast. Generating good dark field conditions when inspecting a curved optical, i.e. specular, surface is not trivial. Using a multitude of illumination angles is necessary for good (and orientation-independent) imperfection visibility, but every additional illumination direction increases the risk of specular reflections into the camera, leading to blind spots in the images. We have developed two approaches to deal with this problem: In the first, reflections are avoided by design, in the second, reflections are treated in data processing. Our commercially available ARGOS system is using a line-scan camera in combination with a rotation stage. The light source is illuminating from almost every direction, except for the direction parallel to the line sensor of the camera itself. This gives excellent imperfection visibility while avoiding direct reflections and maintaining dark-field conditions on rotationally symmetric samples with almost arbitrary curvature. The second approach that has been successfully implemented is suitable also for matrix cameras: Multiple illumination configurations are used sequentially, each generating direct reflections in different portions of the image. These image-sets are then combined into a dark-field image without reflections, by applying suitable filter masks to remove specular reflections from the images. The ARGOS-approach with the line-scan camera has the additional advantage, that very high image resolution is possible by rotating the sample while observing a line aligned with the sample radius. In the standard version, image resolution is up to 250 mega pixels; in the high-resolution version up to 1 giga pixel is possible – in scan times of a few seconds. On the other hand, the requirement to rotate the sample is not ideal for an automated inspection of a batch of samples. In cases when the large resolution is not required, e.g. for smaller elements, using a matrix camera is more efficient, as the acquisition of a sufficient number of images with different illumination conditions takes only a fraction of a second, allowing for short overall inspection times.

Abstract Interior rooms, illuminated by daylight are considered ideal in contemporary architecture. Actual results have been less than perfect as of yet. Light gathering optics and quartz glass fibers make it achievable for the first time.

The testing of highly aspheric optics often requires complex test arrangements: these test systems can be multi-element and will have both fabrication and alignment errors present in the test wavefront. It may not be feasible to calibrate such systems with conventional optical shop practice. The use of diffractive imitator optics, with carefully controlled fabrication uncertainties, can be used to characterise these systems. We describe the use of reflective imitator CGH optics as calibration artefacts in the calibration of an optical test system used to test ELT primary mirror segments. The optical test system is designed to have two operational modes: one to measure a spherical reference optic; and one to measure the primary mirror segment. The use of diffractive imitators in this test system is designed to provide traceability between these two operational configurations, to quantify residual alignment aberrations, and to quantify fabrication errors in the test system. We outline the design of the optical test system, the design of three imitator CGH artefacts required to provide traceability between the two optical test modes, and our calibration approach. We demonstrate the calibration performance achieved with this approach. Without the use of these imitator artefacts, the absolute accuracy of the optical test is estimated to be 149 nm RMS wavefront, of which 47 nm RMS is attributed to midspatial wavefront errors and 141 nm RMS is attributed to alignment and prescription errors. The repeatability of this calibration has been established as better than 3 nm wavefront standard deviation, with an absolute accuracy of 19 nm RMS wavefront.

Measurements of concentration of sensitizers for photodynamic therapy can provide important information in the dosimetry planning and can also give input to the optimal time for treatment. There has been skepticism towards fluorescence techniques for this purpose, as the signal depends on the fluorescence yield and optical properties of the tissue. Absorption based techniques, lack on the other hand, often the sensitivity required for many sensitizers with relative weak absorption in a wavelength region where hemoglobin absorption is dominant. A direct comparison between absorption and fluorescence techniques for measuring mTHPC concentration after topical application on hairless SKH-1 mice bearing skin carcinomas has been performed. 20 μl/cm2 of m-THPC thermogel (0.5 mg m-THPC/ml) were applied on normal and tumor area and the concentration of mTHPC was measured at 4 and 6 hours after drug application by two methods: 1. A fluorescence imaging system capturing images at two wavelengths (500 and 650 nm) following 405 nm excitation. Signals from different regions of interest were averaged and the intensity ratio at 650 to 500 was calculated. 2. A diffuse reflectance spectroscopy system with a fiber separation of 2 mm, providing the absorbance at 652 nm. Both systems provided consistent results related to the photosensitizer concentration. The methods show a remarkable difference in the concentration of photosensitizer in normal skin and tumor. No significant difference in mTHPC concentration in tumor could be observed between the 4 and 6h groups after drug application.

We report on applications of Yb³⁺-doped fiber-laser systems in the micro-processing and the markign of different materials. We have demonstrated the drilling of small holes in glass slides with a thickness of 1 mm using a fiber-laser system wtih an otuput-power of about 35 watts. Furthermore, we have used a narrowband fiber-laser system with an output-power of a few watts for the marking of thin plastic films and samples of anodized aluminum. These applications are based on the experience of more than 10 years in the design, development and preparation of rare-earth doped double-clad silica fibers at the IPHT e.V. Jena. We systematically have optimized the composition and the geometry of these specialty fibers. So we could manufacture very efficient samples which were tested in several fiber-laser setups with output-powers of up to almost 200 watts. Especialy, in the high-power region we could solve some thermal problems using new concepts for the double-clad fiber cross section. We successfully designed, prepared and tested so-called Large-Mode-Area fibers as well as fibers with two or mroe refractive indices in the pump-core. Additionally, we increased the available pump-power of the one-end-pumped fiber-laser systems using a wavelength-multiplexed diode-laser fiber-coupled system.

The availability of ultra short (ps and sub-ps) pulsed lasers has stimulated a growing interest in exploiting the enhanced flexibility of femtosecond and/or picosecond laser technology for micro-machining. The high peak powers available at relatively low single pulse energies potentially allow for a precise localization of photon energy, either on the surface or inside (transparent) materials. Three dimensional micro structuring of bulk transparent media without any sign of mechanical cracking has been demonstrated. In this study, the potential of ultra short laser processing was used to modify the cladding-core interface in normal fused silica wave guides. The idea behind this technique is to enforce a local mismatch for total reflection at the interface at minimal mechanic stress. The laser-induced modifications were studied in dependence of pulse width, focal alignment, single pulse energy and pulse overlap. Micro traces with a thickness between 3 and 8 μm were generated with a spacing of 10 μm in the sub-surface region using sub-ps and ps laser pulses at a wavelength of 800 nm. The optical leakage enforced by a micro spiral pattern is significant and can be utilized for medical applications or potentially also for telecommunications and fiber laser technology.

We describe a new method to determine the oxygen saturation and the total hemoglobin content of tissue in vivo absolutely at small source-detector separations (<10 mm). Phase and mean intensity of modulated laser light of various wavelengths was measured at several predetermined source-detector separations in the frequency domain. From these measured quantities, the absorption coefficient was derived using the modified time-integrated microscopic Beer-Lambert law (MBL). In addition, the interaction volume of the photons was determined using a multi-layer Monte-Carlo model of human skin. To evaluate the method, we employed homogenous solid phantoms (consisting of TiO2 particles embedded in resin) with mean scattering and absorbing properties comparable to those of human skin. Furthermore, in vivo measurements were performed in a healthy volunteer to demonstrate that the technique is applicable for the determination of the oxygen saturation and the total hemoglobin content in the skin in vivo. The proposed technique is especially suited for the on-line determination of the oxygen saturation and total hemoglobin content in applications where small applicators are required (e.g., fetal oxygen monitoring sub partu).

An infrared fiber optic sensor has been developed for the in situ detection of chlorinated hydrocarbons and other pollutant species in water. The sensing element consists of a silver halide fiber, coated with an appropriate polymer. The polymer both enriches the chemical species to be measured in the evanescent wave region of the fiber and serves to exclude water from the measurement region. Evanescent wave spectrometry is then used to accurately quantify chemical species such as chlorinated hydrocarbons which have their strongest absorption bands above 10 micrometers . In order to increase the evanescent absorbance signal, and therefore the sensitivity of the sensor, a number of novel launch designs and fiber configurations has been examined. Results from a range of such configurations are presented and conclusions are drawn regarding optimum sensor design.© (1995) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.