RESUMEN
At present, inspection systems process visual data captured by cameras, with deep learning approaches applied to detect defects. Defect detection results usually have an accuracy higher than 94%. Real-life applications, however, are not very common. In this paper, we describe the development of a tire inspection system for the tire industry. We provide methods for processing tire sidewall data obtained from a camera and a laser sensor. The captured data comprise visual and geometric data characterizing the tire surface, providing a real representation of the captured tire sidewall. We use an unfolding process, that is, a polar transform, to further process the camera-obtained data. The principles and automation of the designed polar transform, based on polynomial regression (i.e., supervised learning), are presented. Based on the data from the laser sensor, the detection of abnormalities is performed using an unsupervised clustering method, followed by the classification of defects using the VGG-16 neural network. The inspection system aims to detect trained and untrained abnormalities, namely defects, as opposed to using only supervised learning methods.
Asunto(s)
Aprendizaje Profundo , Aprendizaje Automático no Supervisado , Algoritmos , Análisis por Conglomerados , Redes Neurales de la ComputaciónRESUMEN
This article describes the design of a smart steering wheel intended for use in unobtrusive health and drowsiness monitoring. The aging population, cardiovascular disease, personalized medicine, and driver fatigue were significant motivations for developing a monitoring platform in cars because people spent much time in cars. The purpose was to create a unique, comprehensive monitoring system for the driver. The crucial parameters in health or drowsiness monitoring, such as heart rate, heart rate variability, and blood oxygenation, are measured by an electrocardiograph and oximeter integrated into the steering wheel. In addition, an inertial unit was integrated into the steering wheel to record and analyze the movement patterns performed by the driver while driving. The developed steering wheel was tested under laboratory and real-life conditions. The measured signals were verified by commercial devices to confirm data correctness and accuracy. The resulting signals show the applicability of the developed platform in further detecting specific cardiovascular diseases (especially atrial fibrillation) and drowsiness.