Research On Measuring Method Of Workpiece Size Based On Machine Vision

With the rapid development of modern production,the shape of workpieces in industrial production is increasingly diversified,and measurement of workpieces is an important criterion for judging whether it is qualified or not.Manual measurement can not accurately measure its dimensions.Machine vision measurement and detection technology have the advantages of non-contact,high precision,and high speed.It can solve problems,such as the real-time and stability of size measurement to workpiece,greatly improving the industrial intelligence and automation.This paper takes the elliptical gear as the research object,and uses the machine vision measurement technology to measure it,and designs the workpiece size measurement system based on machine vision.The main research contents are as follows:Firstly,the preprocessing algorithm of workpiece image is studied.Aiming at the deficiency of the median filter algorithm,an adaptive median filter algorithm is designed.Experimental results show that this algorithm has good filtering effect.In the normalization of illumination,an improved illumination normalization algorithm is proposed for the problem that the conventional morphological processing is likely to cause partial edge blurring.Experimental results show that this algorithm improved has good normalization effect.Secondly,the feature extraction algorithm of workpiece image is studied.In the aspect of corner detection,the detection of the top and bottom corners of the workpiece image is completed by the Harris.In image segmentation,by comparing the segmentation effects of OTSU method and fixed threshold method,the OTSU method is used to complete the segmentation of the workpiece image.The experimental results show that the OTSU method can automatically select the optimal threshold value,and the segmentation effect of workpiece image is good.In contour extraction,the contour extraction algorithm based on Snake model can accurately extract the contour of the workpiece and lay the foundation for the next curve fitting.Then,the curve fitting algorithm is studied for the workpiece image.For gear with different shapes,round fitting and ellipse fitting are studied.In terms of circle fitting,the Hough transform algorithm is used to obtain the average value by multiple fittings.The experimental results show that this algorithm can accurately fit the round hole in the gear.In the aspect of ellipse fitting,the tooth tip and the ellipse of the gear can be fitted by the direct least-squares algorithm.Finally,workpiece measurement system based on the machine vision is designed.By the well-built hardware platform and the well-designed software system,the system is tested,and the size of the elliptical gear are measured.The error factors that affect the measurement accuracy are analyzed.