Objective Quantification of Wind Noise

With the rapid growth of electric vehicles (EV), the need for more advanced tools and techniques ever increasing to evaluate wind noise especially in the EV’s due to the absence of masking effect of major noise sources like powertrain. Traditionally, on road evaluation of wind noise is done by either subjectively or objectively using psychoacoustic parameters like Articulation Index (AI) or sound pressure levels. However, the wind noise is usually masked by other noises at low frequency like road noise and at higher frequencies due to different mechanisms like aspiration or leaking, whistling, flow turbulence, cavity and buffeting. This kind of typical behavior of wind noise needs objectification of noise levels across frequency ranges. Since traditional methods lack in characterizing the wind noise frequency wise and brining the human bias especially when dealing with waterfall diagram or color plots so there is a need for development of new method. To avoid the above-mentioned subjective bias this paper deals with the application of colorimetry technique to the color plots measured during the on-road measurements of the vehicle. Colorimetry is a quantification technique for human color perception. In this study, a colorimetric algorithm is employed, where the input is heat-map or colour plots of frequencies obtained from any frequency measuring app, while the output represents the intensity of noise. The characteristics of this heat maps are that colour gradient of the heat map changes according to the noise intensity. This study explores this change in colour gradient. An innovative algorithm is to quantify this noise intensity through rigorous image processing of mentioned heat maps. The intensity value is formulated by extracting RGB values from each pixel of the noise map and transforming these pixel values from RGB space to HSV space. Next, unwanted outlier noises are removed from the noise map, and temporal averaging is applied to the cumulative pixel values. The intensity values obtained from different vehicle wind noise maps are then standardized, and a total intensity value is defined for the comparative study of wind-cutting noise in automotive contexts.