Traditional monocular vision altimetry methods generally have limitations in terms of the camera's setup orientation and lack sufficient research on the height measurement of off-ground objects. To address this issue, an innovative height measurement method is proposed. The proposed method integrates LiDAR point clouds and monocular vision technology, uses geometric imaging as its theoretical foundation, incorporates prior knowledge from the point cloud data of the measured target, and applies geometric transformations to measure the height of objects on arbitrary planes. Additionally, it can accurately monitor the height of objects over time. A Xiaomi smartphone (model Xiaomi 13) was used as the experimental device, and the heights of five prisms were adjusted and measured over three rounds of testing. The experimental results indicate that the maximum relative error for a single prism is 6.559%, with an average error of 4.064%. For other target objects in the experimental environment, the height measurement accuracy reaches 98.043%.