Poster + Paper
20 November 2024 A computer vision system for measuring the geometric parameters of helical surfaces obtained by multicoordinate grinding on CNC machines
Author Affiliations +
Conference Poster
Abstract
Helical surfaces are important elements of solid end mills. Their production is carried out using multi-coordinate grinding. Shape errors and reduction in the quality of the screw surface appear due to abrasion, high pressure, and wear of the grinding wheel. Therefore, it is extremely important to measure geometric accuracy, to perform linear and angular measurements, and to study the properties of rake helical surfaces. The paper proposes improvements to monitoring of helical surfaces via the use of a new computer vision system for assessing microtexture on helical surfaces after multiaxis grinding on CNC machines. A computer vision system was developed to evaluate defects on helical surfaces after multi-coordinate grinding on CNC machines, and a comprehensive analysis of existing indicators for recognizing the defect was carried out in a guaranteed range of probability of finding a solution of 99.7% for the distribution density of grinds. To verify the developed method, the accuracy of surfaces obtained by using the one was compared with the measurements carried out using specialized equipment for the control of the accuracy of helical surfaces. A new system for monitoring the accuracy and defects of cutting edges, helical front and rear surfaces allows establishing the main geometric parameters of the cutting edges and cutting wedge such as flute angle and rake angle at the apex using key indicators of the difference in color intensity in the focal zone of the image. When developing this approach, it was found that areas with smaller curvature of the rake surface are more susceptible to the accumulation of helical flute pitch errors after grinding. Experimental studies of the system operation were conducted to provide empirical evidence on helical surfaces after multiaxis grinding on CNC machines, demonstrating excellent convergence and defect recognition accuracy. The accuracy of determining the results of the inclination angles of the microtexture surface after grinding at the control point is 2-2.5 degrees, which allows you to form a comprehensive solution for scanning the surface, which will allow you to apply a simple method of control using a camera in reflected light.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Petr M. Pivkin, Anton M. Yazev, Vladimir A. Grechishnikov, Ekaterina S. Nazarenko, Elena Yu. Malysheva, Lyudmila A. Uvarova, and Alexey B. Nadykto "A computer vision system for measuring the geometric parameters of helical surfaces obtained by multicoordinate grinding on CNC machines", Proc. SPIE 13241, Optical Metrology and Inspection for Industrial Applications XI, 1324124 (20 November 2024); https://doi.org/10.1117/12.3037374
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image processing

Surface finishing

Image segmentation

Computer vision technology

Computing systems

Matrices

Abrasives

RELATED CONTENT

Automatic distribution of vision-tasks on computing clusters
Proceedings of SPIE (January 25 2011)
Real-time crowdedness measuring system for Taejon EXPO '93
Proceedings of SPIE (October 03 1994)
A Computer Vision For Navigation Of Mobile Robots
Proceedings of SPIE (January 01 1987)
Telepresence Systems-Display System Optics
Proceedings of SPIE (March 21 1989)

Back to Top