Many factors have been studied to forecast the production of sugarcane, including the number of stems, stem length, and stem diameter. Although plant height does not directly indicate stem length, determining the plant height can be helpful in cultivation management owing to the close relationship between the two. It is possible to estimate the height of the plant canopy by creating a three-dimensional (3D) model of a field from images taken by a small unmanned aerial vehicle (UAV) equipped with a camera. However, it is difficult to construct a stable 3D model when a canopy is formed by many structures, as in the case of sugarcane plants, which have many thin stems and leaves. In this study, a stable 3D model was constructed from red-green-blue (RGB) images obtained by a UAV in a sugarcane field in the southern part of the Okinawa Prefecture by combining images taken with a camera angled vertically downward with those taken with a camera mounted at a 30° angle from the nadir. Plant height was measured and estimated during the period between planting and harvesting. The plant height obtained from the 3D model had an error of 7.4%, which is an acceptable level of accuracy for cultivation management, because it provides sufficient information on the lodging conditions of sugarcane. By clarifying the distribution of plant height in the entire field, it was possible to confirm the lodging conditions of sugarcane in the field, which is useful information for planning mechanical harvesting.