This paper presents the evaluation of 36 convolutional neural network (CNN) models, which were trained on the same dataset (ImageNet). The aim of this research was to evaluate the performance of pre-trained models on the binary classification of images in a “real-world” application. The classification of wildlife images was the use case, in particular, those of the Eurasian lynx (lat. “Lynx lynx”), which were collected by camera traps in various locations in Croatia. The collected images varied greatly in terms of image quality, while the dataset itself was highly imbalanced in terms of the percentage of images that depicted lynxes.

No abstract

No abstract

No abstract

Target temperature effect on eddy current displacement sensing is modelled, analysed and evaluated by simulation. The equivalent target quality factor is detected as the main factor that, along with the eddy current displacement probe equivalent quality factor, determines this effect. It manifests in ambiguity of displacement measurement, as well as, masking the displacement variation by target temperature variation, and vice versa. The analysis and the simulation show that there is an optimal operating frequency for minimum sensitivity over an acceptable displacement range. The effect can be used for concurrent non-contact estimate of displacement and target temperature with acceptable error less than 5 %.