Rapid method for binary image-based road traffic noise mapping

Abstract

Generally, when generating noise maps, the Weiler-Atherton algorithm is employed for polygon intersection calculations to estimate building obstruction attenuation. However, for larger areas, the computational efficiency of this method heavily depends on computational power and expensive hardware. To address these challenges, our method employs binary images to streamline the calculation of intersection area between the building polygon and the noise propagation path, thereby improving computational efficiency. The proposed method consists of two features: converting polygon shapefiles into binary images and subsequently applying the Bresenham Algorithm to calculate building occlusion attenuation. Our study demonstrated that this approach reduced the calculation time to 39.1 % of the original duration, with a mean absolute discrepancy between the two methods of 0.77 dB(A). By utilizing the proposed method to generate noise maps with comparable accuracy, our approach offers an attractive alternative that reduces hardware requirements.