Digital image processing for aggregate orientation in asphalt concrete mixtures: Discussion

Abstract

The authors have presented an excellent paper on digital image processing for aggregate orientation in asphalt concrete mixtures. They are to be commended on their innovative approach to statistically quantifying the effect of different methods of compaction on the orientation of coarse aggregate particles in different mixtures. The statistical analysis was basically concerned with the orientation of only coarse aggregates which were defined in the paper as those with sizes greater than or equal to 2 mm. Since aggregates are studied in this investigation in light of their orientation, it is therefore imperative to clearly define the criterion used in distinguishing coarse aggregate from the rest. Is the ≥2 mm governing dimension measured on the horizontal or the vertical cross section, or both? In other words, is it necessarily important that aggregates that satisfy the criterion in one direction satisfy it in the second too? If not, will they still be considered acceptable aggregates for analysis purposes? The authorsi thoughts on these questions will be greatly appreciated. The writer was fascinated by the digitalized images shown in Figs. 2 and 3 of the paper. These figures, together with Tables l and 3, make the basis of the following point which the writer wishes to raise. It is clearly stated in the paper that Figs. 2 and 3 are those of a gyratory compacted large stone mix. The authors have selected five large aggregates as outlined in Figs. 3a and 3b, and listed their geometrical properties on the horizontal and vertical cross sections in Table 1. For only five aggregates (Nos. 29, 53, 129, 234, and 246), the total area on the horizontal cross section is 2063.66 mm 2 . For aggregates Nos. 3, 49, 144, 203, and 217, the total area on the vertical cross section is 1479.21 mm 2 . The above areas represent 11.7% and 12.3% of the total cross sections in the horizontal and vertical directions, respectively. Table D1 shows the average of the columns listing the average area percentage and average area per aggregate given in Table 3 of the paper. A comparison of the results indicates that, on the horizontal cross section, only five aggregates represent 11.7% of a total