**4. Discussion**

The standard procedure to obtain illumination levels of a street or road consists of measuring luminance or illuminance with photometers at street level at a few discrete points on the pavement. The standard procedure would be more precise if the number of measurement points was greater. However, this would obviously be a grea<sup>t</sup> deal more work. The methodology proposed (see Figure 15) allowed us to obtain the same parameters but with a greater number of measurement points (all pixels of an image or large surfaces such as roads, streets, etc.). This was accomplished by covering a larger area with a single image captured with an airborne camera. In addition, decision-making capability was improved because geographical information systems were used to implement the levels of lighting and uniformity in addition to the electrical consumption of the installation. This allowed us to calculate the energy efficiency of the installation and its energy class.

**Figure 15.** Diagram of the stages of the methodology. The blue boxes indicate the information provided by the local city council.

In the first example, the proposed method was applied to a small section of a street after calibrating the camera. The only information obtained was the luminance (through each pixel of the image) of the pavement. With these data we were able to calculate illuminance and uniformity. In addition, knowledge of the electrical consumption of the luminaires, the electrical energy lost by the distribution lines (electrical energy lost by the distribution lines is calculated by Boucherot's theorem, where *PLines* = *PTotal* − *PLamps*. The *PLamps* is obtained through the manufacturer's data sheet and *PTotal* through the smart electric meters located in the street light control box) and the area of the surface to be illuminated made it possible to calculate the energy efficiency of the installation. All of this information was used to ascertain whether lighting levels were optimal or whether there was an excessive consumption of electrical energy.

The second example in Figure 12 focused on pedestrian streets and streets with slow traffic in a residential area. Based on these characteristics, we deduced that the lighting class according to the CIE (CIE refers to Commission Internationale de l´Eclairage, in english International Commission of Illumination.) was P [15] (formerly known as S class). More specifically, it was P2, where the recommended illuminance level is 10–15 lx and the recommended uniformity greater than 0.3.

The information provided by the GIS showed that two of the four streets were correctly lit, whereas one of them (street number 2) was overlit. This clearly indicated the need to reduce either the lighting points or the electrical power of the luminaries. Of these two options, the most successful would be to reduce the electrical power of the luminaires, because removing points of light would lessen uniformity. As shown in Figure 14, the uniformity is at the limit of what is recommended by the standard. Still another reason to reduce the electrical power is the energy efficiency of the installation in that street. Figure 16 shows that its efficiency corresponds to energy class D, which means that there is room for improvement.

**Figure 16.** Power consumed in each street (**left**) and energy class (**right**).

In contrast, street number 4 was under illuminated. Figures 13 and 14 show that even though the uniformity is correct, it is very close to the minimum because there is only one luminary on the street. The energy efficiency is low but still acceptable. In this case it would be necessary to redesign the street lighting by adding one more luminaire with slightly lower electrical power. This action would improve the level of lighting and uniformity, while maintaining the same energy classification.
