Nowadays, the developing LiDAR technology has attracted a great deal of attention for generating Digital Terrain Models (DTMs) especially in forest areas, due to great penetration of laser pulses in vegetation. Generating DTM using LiDAR data includes two seteps: filtering and interpolation. Although there is a wide variety of filtering methods, in this study, slope based filter  and hybrid filter have been applied in order to filter the cloud of raw lidar data points.

Then Genetic Algorithm ,for improving the interpolation methods like Inverse Distance Weighted (IDW) and polynomials, have been employed. The performance of these intelligent methods has been compared with typical interpolation methods such as Kriging and Radial Basis Functions, etc.

Results of the present study indicated that interpolating by hybrid filter, which used Genetic Algorithm ,for improving the interpolation  Inverse Distance Weighted(IDW), RMSE of 0.135 m for the first area having a dense vegetation conopy and RMSE of 0.230 m for the second area having grassland vegetation, had the best performance of all the applied interpolation methods in this study.