Development Of An On-machine Inspection System For End Mill Based On Image Processing

With the large-scale application of CNC machining technology in the equipment manufacturing,the pursuit of high-precision,high-efficiency,high-automatic production methods is increasing.As an important part of the machining process system,the cutting tool has great influence on the precision and efficiency of the cutting process,especially in the field of aerospace manufacturing.The complex structure of the aviation components and the high material removal rate result in huge tool usage.At present,the judgment of tool wear state based on manual experience has been unable to meet the needs of highly automated production.Measuring tool geometry parameters and tool wear,is of great significance to improve CNC machining accuracy and automation.Therefore,this paper takes cylindrical end mill as the research object,based on image processing technology,researches and develops a tool wear in-machine detection system,and provides data for the end mill wear research.The main research work of the thesis includes:1)The basic shape and characteristics of tool wear were analyzed.The blunt standard of the tool was introduced,and the flank wear of the turning tool and the end mill was compared.Due to the difficulty in measuring the wear of the flank of the cylindrical end mill,the relationship between the wear of the cutter and the radial dimension of the tool is analyzed from the perspective of the milling micro-element,and the measurement of the wear of the tool is converted into the measurement of the radial dimension of the tool.Through analysis of the geometric characteristics of the end mill surface,the measurement principle of the radial dimension of the end mill is given.2)The radial measurement method of milling cutter based on image processing technology is studied.Based on QLM27100-5X five-axis machining center,the hardware structure and measurement scheme of the measurement system are designed.The performance and requirements of important hardware components were analyzed and selected for system requirements.Including CCD image sensors,embedded processors,A/D converters and parallel light source system.The structure and function modules of the system software are studied,and the main functions of the software are analyzed and the design points are given.With Keil MDK and MATLAB,the program of the system hardware control,image acquisition and transmission,image edge extraction and radial dimension measurement are implemented.3)The image acquisition and image preprocessing scheme of the measurement system was designed and studied.The STM32 microcontroller was used to drive the line array CCD sensor to realize image acquisition and USB transmission.In the filtering noise reduction,the combination of adaptive median filtering algorithm and weighted mean filtering algorithm is used to eliminate data noise.In image segmentation,the results of algorithm based on global threshold and gradient amplitude image is compared and analyzed.The effectiveness of watershed algorithm for image segmentation is verified,and the single pixel edge of the tool image to be tested can be stably extracted.4)The sub-pixel edge detection algorithm is studied,and the sub-pixel edge detection of the tool image is realized to obtain the tool radial dimension.The measurement error of the system is analyzed.The measurement system is calibrated by linear calibration method to improve measurement accuracy of the system,and the feasibility of the measurement system scheme is verified by experiments.