I believe point density is determined by the type of sensor and can only be adjusted prior to the flight. Point density is how many laser pulses are sent out from the sensor. That can be increased or decreased by adjusting the laser settings... Can you clarify what you mean by unrealistic DEM?

i have tested many times different production parameters, but DEM is not showing anything about my study area. DSM also in not uniform, there some mixed pixels between buildings and trees. i also couldn't extract any trees and building from software. Although many times I refined the production parameters, i haven't seen any specific change in the outputs quality. I don't know exactly what the problem is! i think may be the software is not using all lidar returns.can it be possible? do you think is there any possibility to do any interpolation for point cloud to increase point density?(do you think it is necessary) I really appreciate your help.

Do you mean that the point density is 0.8 points per square meter? In that case, E3De will not look for buildings or trees when processing the data. The E3De Help says: "E3De processes buildings and trees when density is greater than six to eight points per square meter; therefore, it is not recommended to look for buildings and trees in data with less than six points per square meters, as there will likely be too many false readings. The more points per square meter, the more accurate the ability to identify landscape features. For best results, work with high densities up to 50 points per square meter." E3De does not include any tools that artificially increase the point density using interpolation. Basically, that would just be making up data, which is usually not a good idea. So, I think with these data you will not be able to extract buildings or trees. - Peg Exelis VIS

thanks a lot for your useful information.