飞机发动机叶片裂纹自适应检测算法的研究

发动机叶片裂纹检测是机务维护的重要内容,能否快速准确地检测叶片裂纹,对于缩短战场抢修时间、保障飞机的完好率具有重要的意义.本研究采用基于边缘检测的小波滤波和自适应阈值分割算法,实现叶片裂纹的自适应快速检测;基于边缘检测的小波滤波方法能动态的确定分解层次,有效抑制了噪声点的影响,提高了图像的视觉效果;自适应阈值分割是通过迭代的方法寻找图像的最佳分割阈值,能自动、可靠地检测裂纹的位置、长度等特征.仿真结果验证了这种检测算法的有效性,能满足外场对发动机叶片裂纹检测的要求.

Research of the Adaptive Image Processing Arithmetic for Detecting Crack of Airplane Engine Blade

The crack detecting of engine blade is an important content for airplane maintenance. It can not only reduce repairing time but also means great significance for ensuring the perfect of airplane if the crack of engine blade can be detected by quickly and ex- actly. The methods both the removing the noise for wavelet based on the edge detection and the adaptive threshold segmentation a- rithmetic are adopted to achieve the rapidly adaptive crack testing. The method of removing noise for wavelet which is based upon edge detection can dynamically confirm the disassemble level, restrain the image of noise point effectively and improve the vision effect of filter. The adaptive threshold segmentation looks for the optimal threshold segmentation by iterative method. This can reliably detect the deposition and the size of crack by automatic. The simulated result verifies the validity of this testing method. The method can meet with the requirements of outfield for detecting the crack of engine blade.