基于分形关联维数的飞机蒙皮腐蚀特征提取算法

Corrosion feature of aircraft skin extraction algorithm based on correlation dimension of fractal

针对民航飞机维修中的缺陷检测问题,研究了一种基于分形分数维的腐蚀特征提取算法。分析蒙皮的腐蚀现象在分形维数中的定量表示方法,并通过腐蚀区域的轮廓分形特性来确定边界—质心的距离矢量,同时对此矢量进行拟合。在考虑腐蚀区域收缩时阶梯变化的自相似基础上,修正腐蚀轮廓的距离矢量的拟合特征值。把拟合值作为一种关联维数特征,对飞机蒙皮腐蚀缺陷进行检测识别。实验结果表明,该算法对飞机蒙皮腐蚀的特征提取简单,具有腐蚀识别率高、鲁棒性强等优点,且能够满足蒙皮腐蚀缺陷自动检测实时性的要求。

For the defects testing of civil aviation aircraft maintenance, an extraction algorithm of corrosion feature based on fractional dimension of fractal was proposed. Firstly, the ration denotation method of corrosion phenomena in fraetal dimension was studied, and then distance vectors between the boundary and the eentroid were ascertained by fraetal feature of corrosion region＇ s contour and the vectors were fitted. Secondly, the feature value of corrosion contour which results from fitting was modified depending on self-similarity of step variation when the corrosion region was contracted. At last, the fitted value was regarded as a correlation dimension and the corrosion defects of aircraft skin were detected and recognized. The experimental results demonstrate that the proposed algorithm is easy to extract feature of corrosion of aircraft skin, and it has high recognition ratio and robustness for corrosion detection, especially, it can meet the real-time request.