Application Of Machine Vision In Defect Detection Of Mobile Phone Screen And 3D Measurement

With the rapid development of technology and economy,the process of industrialization is also becoming faster.And the appearance of machine vision plays an important role in the development of industrial automation.In this paper,the application of machine vision in the mobile phone screen defect detection and the application in the 3D measurement system were researsched.And Different systems were established for this two applications.Nowadays,in the mobile phone industry,mobile phone production is becoming more and more automated.However,the quality testing process is low efficiency and the detection accuracy is easy to vary.This leads to a lot of waste of manpower and material resources.As a new technology,machine vision use the method of non-contact surface detection,so it has a series of advantages such as high precision,no damage,automation and so on.This paper designs a set of machine vision detection schemes,including image acquisition system,image processing system and software system.And the algorithms used in the detection of glass defects in mobile phone screens were also studied,including image preprocessing,edge detection and template matching.The template matching is combined with the Gaussian pyramid search strategy to extract the screen defects by image difference method.After testing,this method has a high reliability.Machine vision plays an important role in the field of two-dimensional detection and 3D reconstruction.Its application in three-dimensional measurement has also been studied,And the three-dimensional measurement scheme of binocular stereo vision based on structured optical phase shift coding is designed.The phase shift algorithm is combined with the triangulation principle to obtain the three-dimensional information of the object.The camera calibration was studie in this paper,and the calibration algorithm of zhang is improved by using the camera model.The circular calibration plate is used to calibrate and improve the accuracy of the camera calibration.In addition,the paper also studies the phase shift algorithm,designs two sets of projection gratings,uses the multi-frequency heterodyne and the global look-up table to carry on the experiment respectively.Finally,using 12-step phase shift combined with three 5-step phase shift for phase demodulation.In the phase unwinding process,each pixel is encoded along the u direction to get the global code,and create a lookup table.At last,design the entire workflow,and get 3-Dimensional point cloud data.After testing,this method can get high precision point cloud data.