Research Of Location And Inspection Technology Of Plug-in Machine Based On Machine Vision

With the continuous development of electronic assembly technology,more and more higher efficient and precise surface mounting devices have been able to assembly most standardized elements.While there are still some non-standard elements,such as power devices,electronic connectors,plug-in switches,that must apply Through Hole Technology due to the complex structure and diverse shapes.Although China is a big country of electronic product production and consumption,such special-shaped elements are still mainly produced by manual insertion methods,which is in very low efficiency and quality.Based on the above existing situations,it is particularly meaningful and economically valuable to study the Automatic Insertion Technology for non-standard elements.This paper aims to study the key technology of the plug-in machine-Visual Orientation Detection Technology.The main contents are as follows:1.It has an in-depth investigation of both domestic and abroad researching status and function requirements for the automatic plug-in machine,specifying the functions and technical indicators for its visual orientation system,designing its structure accordingly,thus building related experimental platform for lighting system;image collection scheme etc.2.Image preprocessing algorithms such as image filtering,Threshold Segmentation,and edge detection are studied in line with the requirements of Mark point positioning and element feature extraction.An upgraded circle Hough transformation method is also presented by analyzing the common Hough feature extraction algorithm.Through the experimental analysis,the author chooses the image processing and feature extraction algorithm which is most suitable for this system.3.This paper further studies the relevant algorithms of the plug-in machine positioning,including camera calibration;hand-eye calibration;PCB board benchmark locating;and element position rectification.By camera calibration and hand-eye calibration,it puts forward a conversion method for image coordinate and world coordinate,together with a vector-based element-rectification algorithm to solve the problem of element position deviation.4.By using C ++ programming language to realize algorithms and software of visual orientation system under VS2012,this paper shows different experiments to complete the verification and analysis of fixed camera calibration;the arm camera calibration;PCB board benchmark locating;and element position rectification.