In the quasistatic limit, a time-varying magnetic field inside a conductor is governed by the diffusion equation. Despite the occurrence of this scenario in many popular physics demonstrations, the concept of magnetic diffusion appears not to have garnered much attention itself as a subject for teaching. We employ the model of a thin conducting tube in a time-varying axial field to introduce magnetic diffusion and connect it to the related phenomenon of inductive shielding. We describe a very simple apparatus utilizing a wide-band Hall-effect sensor to measure these effects with a variety of samples. Quantitative results for diffusion time constants and shielding cutoff frequencies are consistent with a single, sample-specific parameter given by the product of the tube radius, thickness, and electrical conductivity. The Laplace transform arises naturally in regard to the time and frequency domain solutions presented here, and the utility of the technique is highlighted in several places.

IntroductionMachine learning tasks often require a significant amount of training data for the resultant network to perform suitably for a given problem in any domain. In agriculture, dataset sizes are further limited by phenotypical differences between two plants of the same genotype, often as a result of different growing conditions. Synthetically-augmented datasets have shown promise in improving existing models when real data is not available.MethodsIn this paper, we employ a contrastive unpaired translation (CUT) generative adversarial network (GAN) and simple image processing techniques to translate indoor plant images to appear as field images. While we train our network to translate an image containing only a single plant, we show that our method is easily extendable to produce multiple-plant field images.ResultsFurthermore, we use our synthetic multi-plant images to train several YoloV5 nano object detection models to perform the task of plant detection and measure the accuracy of the model on real field data images.DiscussionThe inclusion of training data generated by the CUT-GAN leads to better plant detection performance compared to a network trained solely on real data.