Coarse-grained granular soils are critical construction and building materials. The particle shape of granular soils significantly affects their macro mechanical properties. The particle shape can be quantified by sphericity, roundness, and roughness at three different scales. At each scale, researchers have proposed various descriptors for shape characterizations. Before the advent of computers and digital cameras, the two-dimensional (2D) maximum particle projections were utilized to determine particle shape descriptors through either manual measurements or visual comparisons. In the last three decades, image-based techniques have been utilized to automate particle shape analysis. However, there is a lack of comprehensive review of existing particle shape definitions and image-based techniques for particle shape analysis. As such, this study first investigates the particle shape descriptions and image acquisition system. The image-based methods for computed-aided particle shape analysis and influence factors are also reviewed. Further, the review emphasizes empirical relationship between particle shape and mechanical properties of granular soil for engineering applications. The results indicate that different descriptors yield inconsistent outcomes. Therefore, researchers should choose appropriate particle shape descriptors and evaluation techniques based on the specific problems that need to be addressed.
