Key problem in processing data originating from airborne laser scanning is filtration understood as a procedure of elimination of all points which do not belong to specific surface in measured cloud of points. It enables determination of topographic surface course, that is determine DTM. Preliminary classification is usually performed automatically with use of various algorithms for automatical filtration of laser scanning data. Depending on used program exist possibilities of such filtration correction through manual classification. In the paper analysis of automats applied in classification of cloud of points originating from airborne laser scanning on example of TerraScan program. Results obtained using automates with different parameters were compared, and the most often appearing errors DTM and DSM were taken into consideration. Investigations were performed on data originating from the airborne laser scanning performed in 2011, containing a part of Silesia Voivodship. The scanning was performed from a plane using LMS-Q680i scanner with the average height of flight about 850 meters, and the flight was proceeded in two directions: north-south, and east-west with transverse scanning angle respectively 25 and 30 grades. Average density of scanning in single flight came to 6 points on a square meter. ...
