This study verified whether drone-captured remote images can achieve judgment accuracy equivalent to or exceeding that of visual inspection by human inspectors（conducted by specialized inspectors）at large amusement facilities. This addresses challenges in ensuring inspector safety, improving accuracy, and reducing costs and time associated with current manual inspections. The proof-of-concept experiment was conducted by conｆirming resolution and eld of view using simulated deterioration samples, and by comparing manual inspections with automated inspections on an actual roller coaster and Ferris wheel. The results showed that while inspection time effciency has not yet been achieved, drone inspections are useful for improving inspector safety and inspection accuracy. They also offer a significant cost advantage by eliminating the need for aerial work platforms.

View full abstract