Research On Automatic Assembly Of Mobile Fingerprint Module Based On Visual Guidance

The mobile phone fingerprint module is one of the most widely used examples of fingerprint technology.Its main functions include automatic payment,unlocking screen and security management.At present,the fingerprint mode assembly mainly relies on the naked eye detection of the glue road of mobile phone shell before assembling fingerprint module and mobile phone shell.This is not only inefficient but also easy to scratch the product resulting in poor quality.With the rapid development of the industrial robot assembly technology and machine vision technology,the use of automatic assembly system instead of manual operation can effectively solve the above deficiencies.Therefore,the subject of the fingerprint module assembly system is studied;the scheme of the fingerprint module assembly system is proposed;the motion control system is established;the relevant image processing algorithm is designed;and the experimental analysis is carried out.According to the function requirement of the fingerprint module automatic assembly,a scheme of automatic assembly system based on machine vision is proposed.According to the technical requirements and functions of automatic assembly of fingerprint module,the assembly process flow is designed,and the overall structure scheme of the system is determined.Design the structure of the load shifting module and the bending module,and build the visual detection platform to ensure image collection.According to the functional requirements of the motion control system,the plan of using PLC and RC90 as the main motion controller was determined;the hardware selection of the motion control of the load shifting and bending modules was completed;and the motion control process was designed.The I/O communication between robot and PLC and TCP/IP communication between robot and vision detection system are presented.In order to ensure the accuracy of visual detection,the manual calibration of industrial camera and robot was carried out.For AOI detection before assembly,the glue path is positioned through image filtering and improved template matching,and the refinement of the glue path is completed by affine transformation,and then the connected domain area is calculated to complete the glue path detection.Assembly correction is to use the improved template matching algorithm to complete the positioning of mobile phone shell,and the fingerprint module area is extracted through the grayscale transformation and the connected domain mark.After using the Sobel operator for edge detection,thecontour segmentation is completed through the Ramer operator to solve the extraction problem of feature circular arc segment.Finally,the fingerprint module positioning is performed by fitting the circle by the least square method,and the industrial robot is corrected to achieve accurate assembly.Based on the above work,the verification test related to the fingerprint module assembly system was conducted,including the hand-eye calibration experiment and precision analysis,the repeated positioning accuracy of the key axis and the experimental test analysis of the automatic assembly system image processing.This test can explain the automatic assembly system is able to meet the demands.