Study On State Recognition Of Quick Milling Cutter Based On Machine Vision

Milling cutter is widely used in the field of mechanical processing.Quick milling cutter,the wear is more severe in the process of high-speed milling,which wear rate is faster than that of medium-low-speed milling,and its wear capacity is large.However,degree of milling cutter wear is directly related to product processing quality and efficiency,and even impact on the life of machine tools.There is higher requirement for the machining accuracy of products in mechanical manufacturing with progress and development of science and technology.It is of great significance to grasp wear state of milling cutter in time for improving production efficiency and quality.In order to ensure machining accuracy and quality of work piece,and further improve the automation degree of NC milling process,this paper regards wear state of quick milling cutter as research object,after investigating and researching the tool detection technology,which acquire following exploratory work that direct at detection method of machine vision system is undertaken:The wear characteristics and process of the tool was described in detail after comparing and analyzing two different methods including traditional tool wear detection and machine vision detection.Putting fully understanding of causes of tool wear as a prerequisite,characteristics and mechanism of the quick milling cutter wear was explained emphatically.According to the wear characteristics and rules of quick milling cutter,standard of tool bluntness used by this paper was worked out with researching on standard of tool bluntness.Machine vision based design on the detection platform was completed.With introduction of hardware structure of machine vision and technical analysis,selection for sundry machine vision equipment was determined.Meanwhile,design of the milling cutter clamping equipment matched with machine vision experimental platform used by this paper was completed.Consequently,image acquisition was realized successfully.A series of experiments was made on the defect image of milling cutter collected in this paper with carrying out the research of image processing technology.Finally,the image pre-processing from automatic gray adjustment,adaptive median filtering and automatic threshold segmentation to edge extraction of the Sobel operator was obtained.Then,a feature extraction algorithm for this paper was worked out after analyzing and judging.Image processing software was designed and exploited.Software of image processing was compiled and debugged by using MATLAB GUI,and functions of two portions including control part of the milling cutter clamping equipment and image acquisition part were integrated into the software in order to realize the automatic detection of the machine vision inspection system.Accroding to comparing the experimental methods,it was proved that the machine vision measurement method is efficient and accurate by using different methods of milling cutter wear measurement.