| With the development of industrial automation, vision measurement is more widelyused in industrial production for its advantage of non-contact, real-time, high reliability andconvenience, etc. To meet the dedicated user’s requirement, the development with highspecific image measurement and analysis system becomes necessary, which has the abilityto provide secondary development interface and meet the requirements of professionalfeedback control.In order to ensure the effective accuracy of the workpiece, the Image measuringinstrument measures the wear condition of the cutting tool of the mobile phone screenglassindirectly, and then compensate for the wear. It measures the size of the glass, andcompares it with the standard size. The result is used to figure out the wear condition of thetool.Visual measurement extracts the image edge information by the technology of imageprocessing, to identify the image shape and measure the size. In this paper, the digital imageprocessing technology is introduced detailedly, and some improvements methods arepresented. The improved gray-scale combined with the number of times of the gray-scalechanges is chosen as the definition evaluation function, the hill-climbing algorithm forauto-focusing. The Roberts operator is used to locate the coarse edge for its high speed incalculation and thin edge extraction, and the linear interpolation method to detect thesubpixel edge. This paper also presents a method to split multiple graphics area based onedge detection, and identify the corners of the outline curve. Then the improved leastsquares method is used for curve fitting, and the measurement accuracy better than3micron.Finally, a large number of experiments are carried out, the error analyzed, and methods oferror reducing and error compensation are presented.The paper also introduces the hardware system of the image measuring instrumentbriefly. The Glade software is adopted to design the interactive interface, which isuser-friendly and easy to operate. |
PDF Full Text Request