Estimating lighting device inventories with the LANcube v2 multiangular radiometer.

This paper is a technical report describing the use of an open-source instrument called LANcube v2 to estimate a lighting devices inventory. The instrument has 5 color sensitive sensors, each on a face of a cube. The instrument can be mounted on a car roof to create a map of the artificial light at night while roaming the streets and roads. Based on the temporal variations of the detected signal on various cube’s faces, we developed a method of finding the approximate position in 3D of each source. The lamp spectral types can be determined relatively well thanks to the color balance of the raw Red (R), Green (G), Blue (B), and Clear (C) color bands. If one assumes a typical angular photometry of a source with respect to its location, it is possible to estimate roughly its current luminous flux. Such information allows us to build a lighting devices inventory of a territory. One advantage of that new method is that it can provide information about the private sources that are always excluded from public lighting inventories. We compared the inventory extracted with that new methodology with an in-situ lamp inventory made for two villages in Canada. This comparison allows us to emphasize the strengths and limitations of the method by comparing to the ground truth. We found that we were able to detect 99% of the sources with flux higher than 1000 lumen and located within 15 meters from the road. We also found that we generally overestimate the height of the devices by 21 to 51% depending on the lamp photometry. This overestimate surely reflects in an overestimate of the fluxes. Finally, we found that the proposed method is very efficient to recognize the spectral type of the devices with 99% of success. [ABSTRACT FROM AUTHOR]