Resumen

MEJIA U., Santiago M.; GARCIA F., Willman  y  BRUSCHETTA, F.. Detection system and permanent monitoring by image of fires and air pollution in the city of Cochabamba. RevActaNova. [online]. 2017, vol.8, n.2, pp. 216-228. ISSN 1683-0789.

This work demonstrates the feasibility of using a system of real time detection and monitoring of fires and air pollution in Cochabamba based on an image analysis. Using low cost electronic components and adequate image processing, it is possible to detect fires in real time with a high efficiency percentage, while analyzing the image for contaminated sectors and correlating the data permits to obtain an approximation of the level of Pollution through digital photography. The percentages of fire detection efficiency at distances up to 1.5 km is in the range from 60% to 85%, up to 3 km from 50% to 70%, and up to 4 km from 40% to 60%; most of detection errors is related to light errors and to small fire pockets. According to the terrain topography the amount of light errors can increase if there is an important section of flat earth in the image, which clearly reflects the passage of clouds. The smog detection results show that the level of contamination is observable within the height of the blend layer, with a marked gray tendency in the range of 170 (scale 0 to 255). The variety of camera implementations that may be compatible with the Raspberry Pi plate provides the opportunity to improve efficiency percentages.

Palabras llave : fire; air pollution; detection; monitoring; real-time.