A method and apparatus for determining ground clearance of a structure are provided. The method includes removing reflection points caused by noise from first reflection point information based on temporal persistency and generating second reflection point information, extracting visual features from an image of a camera based on convolutional neural network, projecting the second reflection point information onto the image, generating region proposals based on the projected second reflection point information and the image, the region proposals indicating potential horizontal structures above a path, detecting stationary horizontal structure above a path based on the generated region proposals, and determining distance between ground and the detected stationary horizontal structure based on the projected reflection point information and the image.