Colour-histogram-based assessment procedure of the rolling bottle test considering the drawbacks in the digital image analysis of bitumen–aggregate systems

Abstract

The paper presents the results of an investigation that focused on the influence of lighting conditions and camera management (the attaching mode, angle of scanning, and distance of camera lens) on the chromaticity (the range of shades) of an image of the aggregate. No significant differences in the range of shades were revealed in the aggregate (andesite and limestone) and bitumen that were placed on two backgrounds (black and white paper). The possibility of distinguishing the aggregate and bitumen was assessed using their range of shades in a brightness histogram. The overlapping of the ranges of the aggregate and background shades was identified. Similarly, it was identified that the ranges of the aggregate and bitumen overlapped. The percentage of errors was different. The errors that occurred when the white background and dry surface of the aggregate were assessed were minimal and acceptable. A verification of the proposed methodology that focused on distinguishing the aggregate and bitumen via the colour histogram proved that the best method for determination of the degree of bitumen coverage was by using the average of the values determined for each aggregate particle. This method was more time consuming, but it eliminated the errors due to the shadows on the background caused by the aggregate particles arranged with gaps. It also eliminated the overestimation of the degree of bitumen coverage owing to the dark areas among the aggregate particles stored without gaps.