Abstract: To establish an accurate fine mechanical model of asphalt mixtures and optimize the performance of mixtures, it is necessary to construct an aggregate model with real morphological characteristics and calculate and analyze the morphological parameters of aggregates. In this paper, the three-dimensional (3D) laser scanning technology was used to collect the 3D point cloud data of 500 samples of four kinds of aggregates, namely pyroxene, limestone, basalt and pebble, to simplify the aggregate model through the spherical harmonic function, and to establish a 3D aggregate sample library. The morphology indexes of existing aggregates were summarized, and F_I and A_I were selected to quantitatively characterize the 3D morphology of coarse aggregates through Pearson correlation analysis, and F_I and A_I were statistically analyzed for different types and grain sizes of aggregates. Results show that the basic shape, volume and surface area of the simplified aggregate model are very close to those of the real aggregates when the order of the spherical harmonic series is 10 and the number of triangular mapping grids is 643; the F_I and A_I of aggregates obey the normal and lognormal distributions, and varies with aggregate types and grain sizes.