CN 113744230 A

Abstract

一种基于无人机视觉的飞机蒙皮损伤智能检测方法。其包括构建原始飞机蒙皮图像样本数据集；构建高质量飞机蒙皮图像样本数据集；构建飞机蒙皮损伤图像样本数据集；构建飞机蒙皮表面图像采集系统；规划无人机绕检方案；获得飞机蒙皮损伤待测数据集；获得训练好的飞机蒙皮损伤检测网络模型；利用训练好的飞机蒙皮损伤检测网络模型对飞机蒙皮损伤待测数据集进行自动检测等步骤。本发明有益效果：基于无人机视觉技术，能够有效地解决传统检测方式在飞机蒙皮损伤检测中出现的检测耗时长、检测精度低、成本费用高等问题，提高了飞机蒙皮无人机绕检技术的精度，进而为民航领域中飞机蒙皮损伤的检测识别提供技术支持。

English (machine translation): An intelligent detection method for aircraft skin damage based on UAV vision. It includes constructing the original aircraft skin image sample dataset; constructing the high-quality aircraft skin image sample dataset; constructing the aircraft skin damage image sample dataset; constructing the aircraft skin surface image acquisition system; planning the UAV circumvention scheme; Obtaining a data set to be tested for aircraft skin damage; obtaining a trained aircraft skin damage detection network model; using the trained aircraft skin damage detection network model to automatically detect the d…

The opening of the abstract. The full text is in the PDF.