Many factors affect the accuracy of the point cloud obtained from terrestrial three-dimentional laser scanning. These factors exist together and influence each other. Therefore, it is useful to effectively improve the accuracy of the point cloud by studying the effect of each factor and their interaction on the number, reflection intensity and standard deviation of point cloud. Set the target color, roughness and distance as the study objects. By using multi-factor variance analysis method and multiple linear regression method, the significance of the influence of the various factors and their interactions are analyzed, and regression equations between the standard deviation and reflection intensity of point cloud are fitted. Results show that distance of target object has a significant impact on the number of point cloud. The number of point cloud at 5 m is about forty times as much as the number at 30 m. Distance is inversely proportional to the number of point cloud. Color of target object has a significant impact on the reflection intensity of point cloud. The strongest reflection intensity of point cloud for white is 0.54, and the one for black is 0.18. The reflection intensity of point cloud from strongest to weakest is white, green, blue, red, black. Distance of the target object has greatest impact on the standard deviation of point cloud obtained from terrestrial three-dimensional laser scanning. The standard deviation of point cloud at 30 m is three times as much as the value at 5 m. And color is taken the second place, however the effect of the roughness is not obvious. There is a power function relation between the standard deviation and reflection intensity of point cloud.