Research On Screw Hole Detection And Lock System Based On Machine Vision

One of the most important processes in the production and assembly process of automatic screw machine is to realize the locking of the screw hole in the product,and to accurately identify the screw hole and accurately locate its center coordinate is the core to complete the locking function of the screw hole.For the screw hole lock with complex product background,the teaching screw machine needs manual identification.The screw hole coordinates are given in advance.The operation steps are cumbersome,the accuracy can not be guaranteed,and the applicable product type is single.Traditional visual screw machine is difficult to distinguish similar interference screw holes correctly,resulting in low accuracy of lock payment and serious mischeck.Therefore,in view of the above problems,this paper takes the middle frame of a certain type of mobile phone as the research object,and designs a set of screw hole detection and lock payment system based on machine vision,to accurately identify and position the screw hole on the middle frame of mobile phone,and finally completes the lock payment function of the screw hole.The main research contents are as follows:First of all,this paper introduces the domestic and foreign research status quo of automatic screw machine and screw machine system,the application of machine vision determines the screw holes detection based on machine vision and lock the overall scheme of the pay system,the system hardware equipment,structures,interface platform and software design,the environment and select the equipment model fit this system.Then,image processing algorithms such as image filtering,image enhancement and image segmentation are studied.By comparing the processing effects of different enhancement and denoising filtering algorithms on the block image of mobile phone,the linear grayscale transformation algorithm is first used to enhance the contrast of screw holes on the frame of mobile phone,and then the filtering algorithm based on the combination of median filtering and mean filtering is used to remove the image noise and smooth the image.Finally,a local dynamic threshold segmentation algorithm based on mean value was used to separate the suspected regions containing all the target screw holes.Secondly,a target detection and recognition algorithm based on regional combination feature and shape template matching is proposed for target screw hole recognition.Feature selection is carried out on suspected regions,and a single rectangular region is generated to match with the target template to realize target screw hole recognition.A positioning algorithm based on sub-pixel edge detection was proposed to locate the target screw hole.The sub-pixel edge contour of the identified screw hole was extracted,and the contour of the extracted feature rules was fitted with the least square method to calculate the mean value of the fitting center coordinate,that is,the center coordinate of the target screw hole.Finally,the visual software system is developed,the human-computer interface is designed,and the system is tested experimentally.The experimental results show that the average accuracy of the proposed algorithm for identifying the screw hole in the mobile phone frame is 93%,and the error of the center coordinate of the screw hole is 0.2mm,which realizes the accurate locking of the screw hole in the frame of the mobile phone under a complicated background,and can basically meet the requirements of industrial production.