Current infrastructure maintenance works face limitations for various reasons: insufficient budget, increasing number of infrastructure facilities requiring maintenance, shortage of labor, and rapidly increasing number of aged infrastructure facilities. To overcome these limitations, a new approach that is different from manual inspection methods under existing rules and regulations is required. In this context, we explored the efficiency of bridge inspection and maintenance using unmanned aerial vehicles (UAVs), which can observe inaccessible areas, be conveniently and easily controlled, and may offer high economic benefits. Various tests were performed on elevated bridges, and suitable UAV images were obtained. The obtained images were inspected using machine vision technology, thereby avoiding subjective evaluations by humans. We also discuss methods for enhancing the objectivity of inspections. Another aim of this study was to automate inspection work and improve work efficiency through computer vision technology. The UAV image analysis and classification technology in this study utilized existing computer vision technology, but the optimization process for each inspection item is described in detail so that it can be directly applied to the inspection task. This is to overcome limitations of current inspection tasks, which require the ability and experience of personnel. For this purpose, objectivity can be secured by optimizing the data acquisition and analysis process on a job-by-job basis. The test results showed that both the efficiency and objectivity of the proposed UAV-based method were superior to those of existing bridge maintenance and inspection methods.