Object detection is recognized as one of the most critical research areas for the perception of self-driving cars. Current vision systems combine visible imaging, LIDAR, and/or RADAR technology, allowing perception of the vehicle’s surroundings. However, harsh weather conditions mitigate the performances of these systems. Under these circumstances, thermal imaging becomes the complementary solution to current systems not only because it makes it possible to detect and recognize the environment in the most extreme conditions, but also because thermal images are compatible with detection and recognition algorithms, such as those based on artificial neural networks. In this paper, an analysis of the resilience of thermal sensors in very unfavorable fog conditions is presented. The goal was to study the operational limits, i.e., the very degraded fog situation beyond which a thermal camera becomes unreliable. For the analysis, the mean pixel intensity and the contrast were used as indicators. Results showed that the angle of view (AOV) of a thermal camera is a determining parameter for object detection in foggy conditions. Additionally, results show that cameras with AOVs 18° and 30° are suitable for object detection, even under thick fog conditions (from 13 m meteorological optical range). These results were extended using object detection software, with which it is shown that, for the pedestrian, a detection rate ≥90% was achieved using the images from the 18° and 30° cameras.

Optimization of lighting installations should be a priority in order to reduce energy consumption and obtrusive light while providing optimal visibility conditions for road users. For the design of lighting installations, it is assumed that the road has homogeneous photometric characteristics and only one viewing angle is used. There are often significant differences between the design of lighting installations and their actual performance. In order to examine whether these differences are due to the photometry of the road, this study proposes metrics to assess the influence of road heterogeneity and observation angle. These metrics have been used on many measurements conducted on site and in the laboratory for different pavements. A calculation engine has been developed to realize road lighting design with several r-tables in the same calculation or for different observation angles. Thus, this study shows that a root mean squared deviation (RMSD) calculation, including average luminance and uniformities associated with different r-tables, is directly correlated to a normalized root mean squared deviation (NRMSD) calculation between these r-tables. With these proposed metrics it is possible to optimize lighting installation while taking into account different types of urban surfaces and the diversity of users.