线程(计算)
螺纹
自动X射线检查
计量学
计算机视觉
机器视觉
生产线
计量系统
像素
计算机科学
人工智能
工程类
光学
图像处理
声学
机械工程
物理
图像(数学)
天文
作者
Randal J. Hunsicker,John Patten,Alton Ledford,Cathie Ferman,Michael S. Allen,Clark Ellis
标识
DOI:10.1016/0278-6125(94)p2586-4
摘要
Production of threaded parts in some manufacturing environments requires high-precision thread measurement and a reliable inspection method. This paper describes considerations and techniques used in the development of a vision system for measurement and inspection of male screw threads. A two-stage process was developed to calibrate the imaging system. This procedure uses a calibration gauge pin to calibrate the vision system's pixel scale and two pitch standards (max, min) to correct for lighting and image distortion. New and enhanced imaging techniques were used to determine outer and inner diameters for each thread pair. These techniques enable the system to determine the maximum, minimum, and average inner and outer diameters of each part. Pitch measurements were obtained using cross sections of each thread pair in the image. The dark and light (crest and root) areas were examined to deduce pitch diameter very precisely. Measurement techniques for overall thread length and full thread length were found to be beneficial to the part inspection process. Lab results as well as production line measurement results using the vision system are compared to optical comparator, micrometer, and gauge measurements. Overall accuracy was within ± 0.0005″ of gauge measurements. Some measurements exceeded this accuracy in lab tests and on the production line. The thread measurement system was demonstrated to be extremely beneficial during the manufacturing process.
科研通智能强力驱动
Strongly Powered by AbleSci AI