Indoor three-dimensional raster map feature point extraction method based on sliding window

The present invention discloses an indoor three-dimensional raster map feature point extraction method based on a sliding window. The method comprises the following steps: step 1, using a graph-SLAM algorithm to obtain a three-dimensional raster map of an indoor environment; step 2, establishing a sliding window, and calculating a duty ratio; step 3, moving the sliding window of step 2 in any direction, wherein if duty ratios all change obviously, the sliding window is a primary feature point; step 4, performing plane segmentation on the three-dimensional raster map of step 1 by using a methodbased on region growing; step 5, performing plane fitting on the three-dimensional raster map that undergoes segmentation in step 4; step 6, calculating distances from the primary feature point of step 3 to three planes adjacent to the primary feature point, wherein if the distances are less than a threshold, the primary feature point is a correct feature point; and step 7, performing aggregationon the correct feature po