公转滑动轴承外圈划伤故障图像特征提取方法

Image feature extraction method for scratch fault of the outer ring of revolution sliding bearing

公转滑动轴承划伤缺口较小,无法准确定位划伤位置,在公转滑动轴承划伤状态下长时间运行对矿用螺旋清仓机的安全产生直接影响,因此,提出矿用螺旋清仓机公转滑动轴承外圈划伤故障图像特征提取方法。采用边缘相似度非局部均值去噪方法对轴承图像展开去噪处理,消除图像中存在的噪声,凸显图像细节信息;对去噪后的图像展开二值化处理,以此突出划伤故障特征,在模板匹配的基础上确定滑动轴承的划伤区域;引入灰度共生矩阵提取划伤故障特征,建立划伤故障特征集,通过特征向量选择算法确定最终的轴承外圈划伤故障特征。实验结果表明,所提方法去噪后的划伤故障图像清晰度高、划伤故障特征提取精度高、提取效率高。

The scratch notch of the rotary sliding bearing is small,and the scratch position cannot be accurately located.Long time operation under the scratch condition of the rotary sliding bearing has a direct impact on the safety of the mining spiral bin cleaning machine.Therefore,a feature extraction method for the scratch fault image of the rotary sliding bearing outer ring of the mining spiral bin cleaning machine is proposed.The edge similarity non local mean denoising method is used to denoise the bearing image,eliminate the noise in the image and highlight the image details.The image after denoising is binarized to highlight the scratch fault characteristics,and the scratch area of sliding bearing is determined based on template matching.The gray level co-occurrence matrix is introduced to extract the scratch fault features,and the scratch fault feature set is established.The final bearing outer ring scratch fault features are determined through the feature vector selection algorithm.The experimental results show that the proposed method has high definition,high precision and high efficiency of scratch fault feature extraction.