Research And Application Of Knife Nut Sorting System Based On Machine Vision

In recent years,with the rapid development of domestic science and technology,the requirements for quality inspection in industrial production are getting higher and higher and traditional inspection methods can no longer meet the requirements of high-precision inspection.Knurled copper nut is a kind of high precision part,often used in fine mechanical structure.At present,most of the chaotic knurled copper nuts still remain in manual sorting,and are classified by measuring the specified index of the nut.This method has been unable to adapt to today's high-quality and efficient industrial production models.It is not only easily to cause false detection but also the detection efficiency is low.Knurling nut sorting system based on machine vision proposed in this paper,because machine vision measurement is non-contact measurement and has high efficiency.In consideration of the structural characteristics of the knurled nut,the sorting system detects by measuring the height,diameter,and number of threads of the knurled nut.Among them,the height and diameter measured by visual measurement method and the traditional image segmentation algorithm for thread number detection.The design of the system is carried out through three aspects:hardware selection,software programming,and image algorithm.Hardware selection part includes camera,lens,light source and motion control card selection,etc.Software programming part includes user interface design and system software code programming.The system uses the C#language and Halcon software for programming.The algorithm improvement part intends to propose a subpixel-based fit circle algorithm and a fit line algorithm to improve the accuracy and efficiency of nut height and diameter measurement.At the end of the paper,it conducted online real-time testing of algorithms and systems The test results show that it can quickly and accurately sort the knurled nuts.