Research On In-situ Detection Technology Of Micro-milling Tool Wear Based On Machine Vision

In the process of micro milling,the wear state of the tool directly affects the quality of the machining surface and the machining accuracy.Excessive wear or damage of the tool will significantly reduce the dimensional accuracy of the workpiece,and even cause the workpiece to be scrapped.Therefore,it is very important to grasp the wear state of the tool and timely change or stop the tool.In view of the shortcomings of offline tool wear detection and existing in-situ detection methods,this paper uses machine vision detection methods to study the in-situ detection technology of micro-milling tool wear.In this paper,the wear patterns of micro milling tools are analyzed and summarized.According to the selected tool wear evaluation index,the hardware of machine vision system is selected and calculated in detail,and the micro-milling tool wear in-situ detection system is built according to the overall design.In order to solve the problem that it is difficult to get the clear focus image accurately through the subjective focus of the naked eye in the micro milling tool wear in-situ detection,the sharpness evaluation of the micro milling tool wear detection image is studied.For some shortcomings of the traditional sharpness evaluation function,a sharpness evaluation algorithm based on the local variance information entropy is proposed.Experimental results show that the algorithm can take into account both high sensitivity and noise immunity,and is superior to traditional definition evaluation functions.In order to solve the problem that the depth of field of the lens may be limited in the wear detection of the micro-milling tool,the tool wear image cannot be clearly focused,the extend depth of field technology in the wear detection of the micro-milling tool is studied.Through experimental simulations,the extend depth of field algorithm based on wavelet transform used in this paper has a better extend depth of field effect and can obtain all clear focused images of high quality of the tool.Processing of high-quality tool wear images obtained through autofocus or extend depth of field.In the aspect image preprocessing,median filtering is used to denoise the image,and in the aspect image segmentation,region growing method is used to get better segmentation effect.Aiming at the different orientations of the images collected during in-situ detection,a tool rotation and positioning algorithm based on Hough transform was studied.In the aspect of feature extraction,the edge of wear area of corner is reconstructed by scanning and positioning feature points and least square line fitting.The maximum wear width,wear area and the reduction of bottom edge diameter are extracted successfully by wear amount extraction algorithm.In-situ detection experiments results show that the detection system in this paper can effectively measure tool wear status and reflect tool life trends.