Metrology plays a critical role in the rapid progress of Artificial Intelligence (AI), particularly in computer vision. This article explores the importance of metrology in image synthesis for computer vision tasks, with a particular focus on object detection for quality control. The aim is to improve the accuracy, reliability and quality of AI models. Through the use of precise measurements, standards and calibration techniques, a carefully constructed dataset has been generated and used to train AI models. By incorporating metrology into AI models, we aim at improving their overall performance and robustness.
The mobility of people and goods is moving into a new era of more automated services based on sensors networks and Artificial Intelligence. At present, Automated Mobility, in the broadest meaning of the terms includes Advanced Driver Assistance System (ADAS) and Autonomous Vehicle (AV), beyond being attractive for many practical advantages, ranging from safety to traffic flow management, still presents several concerns on the trustworthiness of sensor networks integrated into vehicles, especially regarding sensors calibration, data uncertainty and data fusion approaches. Currently, the trustworthiness of ADAS and AV functions is assessed with virtual and physical simulation of functions relying on synthetic sensor models, simulated and measured sensor data and equivalent environmental conditions. During the lifespan of vehicles, environmental effects including possible accidents and common usage, can have significant impact on the performance of ADAS sensors and customer functions. Current approach is to consider ADAS sensors output nominal, disregarding the uncertainty of sensors data and including all the possible tolerances and variability at sensors/system testing stage. At present day, market offers several facilities and commercial set-up promoted as being able to do ADAS calibration: usually are modular equipment allowing alignment and sensitivity check of different ADAS sensors, especially front sensors and camera. However, sensor calibration involves specific calibration facilities, procedures to establishing sensors sensitivity and most of all, associated uncertainty. Accuracy and traceability of ADAS sensors beyond being fundamental requirements in measurement science, allow the quantitative evaluation of the trustworthiness of sensors and customer functions. This paper suggests an approach to lay the foundation of Metrology of Trustworthiness for ADAS and AV complex sensors systems and provide a case study of IMU sensor trustworthiness.