Evidence based on hierarchical fuzzy synthetic lidar data Classification Based

The present invention discloses a LIDAR feature data classification method based on hierarchical fuzzy evidence combination. By constructing the fuzzy model for the first time assigned trust echo image height, the first height difference image last echo, the echo intensity of the first image, the normalized difference vegetation index images respectively assigned trusted, trusted distribution corresponding to an image; using a median filter trust assigned to each image for noise reduction; hierarchical framework structure, synthesizing filtering results for the respective layers, of the synthesis result obtained for the result of classification decisions based on the maximum rule. The present invention overcomes the prior method for high-precision classification slow defect can not meet user needs, while ensuring high accuracy, to effectively enhance the speed of arithmetic operation, a fast and high precision formed feature classification algorithm. It can be applied to three-dimensional modeling of cities, large-scale ecological change assessment, rapid reconnaissance geological disasters and other fields.