The underground separation of gangue from coal is an important part of the process of reducing transportation costs and improving production efficiency. A new method is proposed in this paper to separate gangue from coal on the basis of density, calculated from volume using three-dimensional (3D) laser scanning technology. This approach is based on the laser triangulation method and weight. Taking into account the weigh-in-motion technology that is currently widely applied, the main objective of this approach is to determine the volume of the object being measured. Thus, the principles of 3D laser scanning and laser triangulation were studied in detail and a relative formula was deduced. The reasons and solutions for possible errors are also analyzed in this paper, depending on the method applied to measure volumes of gangue and coal. Physical relationships of objects to be measured, as well as laser and measuring data planes are also presented, depending on principles of 3D laser scanning. A parameter selection method is presented to determine the appropriate photoelectric recognition device to use for coal or gangue, while error due to voids between objects was measured, and transportation belts analyzed. The assumption put forward in this study is that the ratios of voids between objects, transportation belt, and the exact volume of objects all conform to a normal distribution; this was shown to be the case by both experiments and statistical theory. A mathematical model was therefore constructed that is suitable for the recognition of coal and gangue based on this theory, while algorithms of recognition threshold values and identification rate are presented. (C) 2017 Elsevier B.V. All rights reserved.