High-precision sandwich valve outer diameter dynamic visual measurement system and method

Abstract

High-precision sandwich valve outer diameter dynamic visual measurement system, including light source, camera, detection platform and control device, the camera sends the collected photos of the sandwich valve to be tested to the control device, and the control device judges the sandwich valve to be tested according to the photos actual outer diameter. The measurement method of the present invention realizes the dynamic acquisition of the ROI of the sandwich valve contour by adopting the connected domain analysis and the minimum enclosing circle algorithm; adopts the contour detection method to calculate the variance and gradient of the sandwich valve contour points, and removes burr interference; finally adopts the minimum The outer diameter of the sandwich valve is obtained by the square fitting of the edge after removing the interference.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and in particular relates to a high-precision sandwich valve outer diameter dynamic visual measurement system and method. Background technique [0002] As a common rubber seal, the sandwich valve is usually used as a sealing device in the air brake system in the EMU. Its shape is as follows: figure 1 shown. At present, the outer diameter of the sandwich valve is manually measured with a caliper, and then the burr is manually checked visually to judge whether the sandwich valve meets the requirements. However, this method has many disadvantages: the caliper needs to be aligned with the measurement point (the center of the arc of the boss) during manual measurement, which is easy to be improperly operated and cannot be aligned, resulting in wrong detection; the caliper will also have different degrees as the number of uses increases. wear and tear, so that the accuracy of manual ...

AI Technical Summary

Problems solved by technology

However, this method has many disadvantages: the caliper needs to be aligned with the measurement point (the center of the boss arc) during manual measurement, and it is easy to be misaligned due to improper operation, resulting in wrong detection; Wear and tear, so that the accuracy of manual detection cannot be guaranteed (such as figure 2 shown); in addition, the huge labor cost is also a problem that cannot be ignored

Such as "Dai Fengqiang, Liu Tao, Wang Hongbo, Shen Xiaodong. Research on workpiece edge burr detection system based on machine vision [J]. Agricultural Equipment and Vehicle Engineering, 2018, 56(07): 43-47. Tool Technology, 2018, 52(11) :147-149." etc. proposed a method of cumulative probability Hough transform to fit the contour curve, screen and separate the burrs, which can accurately measure the size of the edge burrs of die castings; but the burr detection operator developed by it can only be used for In a few cases, and the requirements for workpiece material and surface technology are high, it is not applicable to the workpiece to be processed by the present invention

Such as "Fu Jun. Design of Automobile Heat Exchanger Dimension Measurement and Appearance Inspection System Based on Machine Vision [D]. Nanjing University of Science and Technology, 2014." A machine vision-based automobile heat exchanger dimension measurement and appearance inspection system was proposed. The system can measure the size and defects of some areas on the automotive heat exchanger; however, the positioning error of the system is relatively large and the measurement algorithm is only tested for repeatability accuracy without actual manual measurement

For example, "Zhao Ping. Research on sand profile measurement system based on machine vision [D]. Shenyang University of Technology, 2013." studied a sand profile measurement system based on machine vision, and the repeatability of the system reached ±0.005mm ; but the experimental system can only be used in a laboratory environment, and its image detection operator cannot eliminate the influence of edge burrs

Images