Ground automatic detecting and scene height calculating method based on depth image

Provided is a ground automatic detecting and scene height calculating method based on a depth image. If h obtained through calculation of a certain point meets the condition that H1ground. All points meeting the condition are extracted, a region with the largest area is found through a binary image labeling method, and the region is identified to be the ground. After the ground is judged, l3 of the ground region in an L_matrix is averaged to obtain indication of a plane normal vector under a camera coordinate system. An average value of the ground region h is the installing height of a camera. For any image point, a height calculation formula is as follows. By means of the method, the depth image is used for conducting ground automatic detecting, and the height between all the points and the ground is calculated in the scene according to a ground detecting result. The height information enables an intelligent monitoring system and an intelligent robot to understand the scene deeply and thoroughly.