A surface relief grating with a period of 30 µm is embossed onto the surface of magnetoactive elastomer (MAE) samples in the presence of a moderate magnetic field of about 180 mT. The grating, which is represented as a set of parallel stripes with two different amplitude reflectivity coefficients, is detected via diffraction of a laser beam in the reflection configuration. Due to the magnetic-field-induced plasticity effect, the grating persists on the MAE surface for at least 90 h if the magnetic field remains present. When the magnetic field is removed, the diffraction efficiency vanishes in a few minutes. The described effect is much more pronounced in MAE samples with larger content of iron filler (80 wt%) than in the samples with lower content of iron filler (70 wt%). A simple theoretical model is proposed to describe the observed dependence of the diffraction efficiency on the applied magnetic field. Possible applications of MAEs as magnetically reconfigurable diffractive optical elements are discussed. It is proposed that the described experimental method can be used as a convenient tool for investigations of the dynamics of magnetically induced plasticity of MAEs on the micrometer scale.
Classification problems are common in Computer Vision. Despite this, there is no dedicated work for the classification of beer bottles. As part of the challenge of the master course Deep Learning, a dataset of 5207 beer bottle images and brand labels was created. An image contains exactly one beer bottle. In this paper we present a deep learning model which classifies pictures of beer bottles in a two step approach. As the first step, a Faster-R-CNN detects image sections relevant for classification independently of the brand. In the second step, the relevant image sections are classified by a ResNet-18. The image section with the highest confidence is returned as class label. We propose a model, with which we surpass the classic one step transfer learning approach and reached an accuracy of 99.86 % during the challenge on the final test dataset. We were able to achieve 100 % accuracy after the challenge ended.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.