刀具磨损自动识别及检测系统

Automatic recognition and detection system for cutter wear

对加工状态下刀具磨损值及时有效的检测,可以在保证加工精度的前提下提高加工效率,同时为刀具剩余寿命预测提供有力的数据支撑。针对刀具磨损值测量过程中人工参与的不足、易受主观因素影响、检测精度低等问题,建立了刀具磨损图像自动检测系统,基于最大类间方差及遗传算法迭代寻找最佳阈值,进行磨损区域分割;利用磨损区域呈现“线状”的特点,进行磨损区域特征识别及滤波;基于多级Hough变换以及磨损区域二次识别的方法,将刀具磨损区域在图像中清晰地提取出来。在此基础上,基于Canny算子边缘检测方法建立刀具磨损曲线,计算出磨损区域的磨损值。在刀具磨损检测系统计算的磨损量与超景深显微镜的人工测量结果进行比对,经实验表明,该磨损检测系统检测误差在±6%以内,并且提高了刀具磨损值的检测效率。

The timely and effective detection of cutter wear value in processing state can improve the machining efficiency on the premise of guaranteeing the machining precision,and provide convincing data support for tool residual life prediction.In view of the defects of manual participation in measuring cutter wear value,affected by subjective factors and the manual detection accuracy was poor,an automatic recognition and detection system was established.Based on the maximum inter-category variance and genetic algorithm,the optimal threshold was searched to segment the wear area.Through the linear feature of the wear area,the wear area was recognized and filtered,which was clearly extracted from the image based on multistage Hough transform and secondary recognition of wear area.On this basis,the cutter wear curve was established based on Canny operator edge detection method,and the wear values of the wear area were calculated.The measured results of the recognition and detection system were compared with those of manual measurement of the ultra-deep microscope,and the results showed that the error between the two is within±6%,which proved that the automatic recognition and detection system for cutter wear improved the efficiency in detecting of cutter wear value.