基于重排Gabor变换和Radon变换的特征提取技术∗

摘要

基于重排Gabor变换和Radon变换理论，提出了基于重排Gabor变换和Radon变换的故障诊断技术。文章首先利用重排Gabor变换时频聚集性特点，对含噪信号进行重排Gabor变换，再将Gabor展开系数进行Radon变换，在Radon变换平面上提取像点，再进行Radon逆变换，实现信号降噪。仿真结果表明，该方法有效地抑制了噪声的干扰，成功地提取了故障特征信号。%According to the reassignment Gabor transform and Radon transform, and a feature extraction method based on the reassignment Gabor transform and Radon transform is proposed. Firstly, the reassign-ment Gabor transform is applied to the noise signal on the basis of its characteristics, and the Gabor coeffi-cients Gabor transform are processed by the Radon transform, and the image point in Radon transform plane is selected, Then the de-noised Gabor coefficients could be obtained by the Radon inverse transform. The simulated results indicate that this method can effectively reduce the noise interference, and the fault features are extracted successfully for mechanical diagnosis.