首页|图像识别在民机机体表面维修中的应用研究

图像识别在民机机体表面维修中的应用研究

扫码查看
在民机机体表面维修领域中,飞机蒙皮损伤检测是非常重要的环节。首先采集飞机蒙皮表面损伤图像,将采集到的图像分为裂纹和腐蚀损伤图像并完成图像预处理工作;为了获得有效准确的特征数据,采用一种结合主成分分析法和形状特征提取的特征提取方法;最后采用支持向量机分类模型进行分类识别,通过选择高斯径向基核函数和合适的惩罚因子,从而对蒙皮损伤类型进行精确分类识别。实验结果表明,该方法能够精确地完成分类识别工作,损伤识别准确率达到95%,满足飞机蒙皮损伤日常检测与维修的需求。
Research on Application of Image Recognition in Surface Maintenance of Civil Aircraft
In the field of civil aircraft body surface maintenance,aircraft skin damage detection is a very important link.Firstly,the image of surface damage of aircraft skin is collected,and the image is divided into crack and corrosion damage image,and the im-age pre-processing is completed,finally,the Support vector machine classification model is used for classification and recognition,and by selecting Gauss's Radial basis function and appropriate punishment factors,the model can be used for classification and recognition,therefore,accurate classification and identification of skin damage types are carried out.The experimental results show that the method can accurately complete the classification and identification work,the accuracy of damage identification reaches 95%.

image processingdamage detection and identificationsupport vector machineairplane skin

杨晓强、谭志强

展开 >

中国民用航空飞行学院航空工程学院,四川 广汉 618307

图像处理 损伤检测与识别 支持向量机 飞机蒙皮

2024

数学的实践与认识
中国科学院数学与系统科学研究院

数学的实践与认识

CSTPCD北大核心
影响因子:0.349
ISSN:1000-0984
年,卷(期):2024.54(3)
  • 14