An intelligent geographic information system-based framework for energy efficient street lighting

An efficient urban development needs to reduce the carbon footprint by providing an innovative solution for street lighting deployment. Hence, a geographic information system is proposed to highlight the illumination levels and energy classes of streets, enabling energy-efficient street lighting. The proposed study investigates the intended area of the satellite image, for initial preprocessing. To identify the closest point the proposed methodology uses the entire training dataset. Subsequently, a classification report comprising the illuminated area and the value of power for each class is obtained. The proposed work is assessed using various performance measures, such as user accuracy, producer accuracy, overall accuracy, and the kappa coefficient. The superiority of the proposed method over traditional techniques shows an improvement in overall accuracy by 17.38-25.38% and the kappa coefficient by increasing its value from 0.2469 to 0.42, respectively.