Research And Application Of Machine Vision Positioning Method In Speaker PCB Mounting

With the popularization of various digital products,the application of electro-acoustic products has become more and more extensive,and the demand has also increased year by year.The speaker is one of the important structures in electro-acoustic components.In the PCB mounting process of speaker assembly,due to the poor flexibility and scalability of current automation equipment,the popularity of automation technology is very low,and PCB mounting of many types of speakers still relies on labor.PCB mounting has higher production requirements,and the manual production method has low efficiency,unstable error and high cost.In view of the above problems,the paper introduces machine vision technology and researches and applies the machine vision positioning method for the PCB mounting process of the headphone speaker.The main research work of the thesis is as follows:(1)The paper introduces the mounting object and mounting requirements,and designs the visual scheme according to the production efficiency and placement accuracy,and expounds the workflow of the visual scheme.Parameter design of the vision equipment in the speaker PCB mounting and equipment selection.Then an experimental platform was set up to collect images of the PCB and the speaker frame.(2)The paper studies the visual positioning method based on image feature extraction.Image pre-processing for noise occurring in PCB and speaker frame images.According to the imaging characteristics of PCB and speaker frame,the positioning method based on edge feature matching is studied.In addition,based on the notch feature of the speaker frame,using image segmentation and morphological operations,a morphological featurebased positioning method is used to locate the speaker frame.Based on the experimental results,the characteristics and applicable conditions of the two positioning methods are summarized.(3)Based on the problem of poor edge characteristics and morphological characteristics of plastic basin frames,the paper uses the deep learning positioning method to conduct related research.Firstly,the image segmentation and edge feature matching method are used to collect and label the dataset of the plastic basin.After that,the design and training of the deep learning algorithm model was carried out,and the model evaluation was carried out in the test set.(4)The paper analyzes the function of PCB mounting,develops the corresponding visual positioning software system for the application of visual positioning method,and introduces the functions and implementation methods of each module in the software system.The system can be used to define and debug the visual device parameter configuration and the visual positioning method,and can realize the switching of different recognition tasks in the production process.It has been verified that the software system can recognize the dispensing track of the speaker frame in the camera field of view,the position of the suction point on the PCB,the position of the placement point on the basin frame and the rotation angle of the manipulator,which meets the positioning requirements in PCB mounting.