From classification results to topographic maps

The classification of high resolution multispectral aerial imagery enables very high thematic accuracies when machine learning methods are applied. The use of classification results as topographic maps requires cartographic enhancement and checking of the geometric accuracy. Urban areas are of special interest. The conversion of the classification result into topographic maps of high thematic and geometric quality is subject of this contribution. After reviewing the existing literature on this topic a methodology is presented. It has the goal to achieve high cartographic quality and geometric accuracy for buildings and other topographic objects. The suggested methodology for improving the classification results is described. With the ISPRS data set of the '2D labelling contest' a land cover map of six classes was produced and then enhanced by the proposed method. The classification used a machine learning method applying a variety of attributes including object heights derived from imagery. The cartographic enhancement is carried out with two different levels of quality. The user's accuracy for the classes "impervious surface" and "building" were above 85% in the 'level 1' map. The geometric accuracy of building corners in the 'level 2' map was assessed by means of reference data derived from a DSM-based orthoimage. The obtained root mean square errors were RMSEₓ = 1. 2 m and RMSEᵧ = 0. 7 m. All processing could be carried out with a high level of automation.