The morphological characteristics of coarse aggregates significantly affect the mechanical properties of asphalt mixtures and thus directly influence their road performance and durability. This paper aims to innovatively propose an evaluation method for the angularity of coarse aggregates. A three-dimensional (3D) scanner was used to acquire the surface profile information of coarse aggregates, and local angularity features were extracted through point cloud data analysis. Subsequently, a comprehensive 3D angularity evaluation index for coarse aggregates (A_3D) was proposed by combining local normal vectors and curvature characteristics. The results indicate that A_3D can be more accurately and independently used for angularity evaluation. The relevant evaluation parameters are as follows: When the value of the basic search radius coefficient μ is 0.05, the intersecting areas of arbitrary fracture surfaces can be effectively extracted; when the dynamic dual-threshold coefficients are 40% < η_k < 50% and 70% < η_σ < 80%, the algorithm has good sensitivity to the variations of curvature and normal vectors and can effectively eliminate redundant data while ensuring the accuracy of feature extraction. Parallel control experiments indicate that when measuring the angularity parameters of the same batch, it is optimal to take the uniform amplification parameter λ as 10 000?15 000. When the content of flaky and elongated particles in coarse aggregates is high, increasing λ to 15 000?20 000 can enhance the ability to distinguish local sharp features. Computational analysis reveals that A_3D is more accurate than the traditional two-dimensional evaluation method, has a good correlation with the three-dimensional voxel angularity index, and shows higher sensitivity to the evaluation indices of flaky and elongated or sharp-edged crushed stone particles. Calculation results at different scaling ratios demonstrate that the algorithm has good adaptability to the particle sizes of gravel and crushed stone coarse aggregates and has better adaptability to the differences in gravel particle sizes.