Professional swimming coaches make use of videos to evaluate their athletes’ performances. Specifically, the videos are manually analyzed in order to observe the movements of all parts of the swimmer’s body during the exercise and to give indications for improving swimming technique. This operation is time-consuming, laborious and error prone. In recent years, alternative technologies have been introduced in the literature, but they still have severe limitations that make their correct and effective use impossible. In fact, the currently available techniques based on image analysis only apply to certain swimming styles; moreover, they are strongly influenced by disturbing elements (i.e., the presence of bubbles, splashes and reflections), resulting in poor measurement accuracy. The use of wearable sensors (accelerometers or photoplethysmographic sensors) or optical markers, although they can guarantee high reliability and accuracy, disturb the performance of the athletes, who tend to dislike these solutions. In this work we introduce swimmerNET, a new marker-less 2D swimmer pose estimation approach based on the combined use of computer vision algorithms and fully convolutional neural networks. By using a single 8 Mpixel wide-angle camera, the proposed system is able to estimate the pose of a swimmer during exercise while guaranteeing adequate measurement accuracy. The method has been successfully tested on several athletes (i.e., different physical characteristics and different swimming technique), obtaining an average error and a standard deviation (worst case scenario for the dataset analyzed) of approximately 1 mm and 10 mm, respectively.

This work presents the development of a laser tracking Doppler vibrometer (TLDV), which is based on a laser scanning vibrometer improved for tracking the same point of a bladed disk during a complete circular rotation, so that a Lagrangian approach can be applied to vibration measurement of a rotating object. The TLDV is presented and applied to the measurement of vibration velocity map of a rotating object. Measurements have been carried out on a bladed disk rotating up to 62 rps. This article discusses the development of the measurement system, with particular attention to the problems of measurement uncertainty and laser spot positioning accuracy. Results of TLDV measurements performed under different conditions of rotation and of vibration are presented. An analytical model of the measurement system was developed in order to evaluate its behavior and analyze its accuracy on the positioning and uncertainty of the measured data. In this way it was possible to identify the most critical elements in the system and to guide its optimization. The misalignment between the optic axis of the tracking system and the rotation axis of the object is the most relevant source of uncertainty, inducing the presence of pseudovibrations at the frequencies of rotation and causing relative motion of the laser beam on the target surface, which turns out to be a positioning error and a source of speckle noise on the laser vibrometer signal. A comparison between static measurement and numerical analysis has been performed, so that capabilities of the tracking technique were highlighted.