Optical flow method is one of the most important methods of analyzing motion images. Optical flow field is used to analyze characteristics of motion objects. According to motion features of micro-electronic mechani-ca...Optical flow method is one of the most important methods of analyzing motion images. Optical flow field is used to analyze characteristics of motion objects. According to motion features of micro-electronic mechani-cal system (MEMS) micro-structure, the optical algorithm based on label field and neighborhood optimization is presented to analyze the in-plane micro-motion of micro-structure. Firstly, high speed motion states for each fre-quency segment of micro-structure in cyclic motion are frozen based on stroboscopic principle. Thus a series of dynamic images of micro-structure are obtained. Secondly, the presented optical algorithm is used to analyze the image sequences, and can obtain reliable and precise optical field and reduce computing time. As micro-resonator of testing object, the phase-amplitude curve of micro-structure is derived. Experimental results indicate that the meas-urement precision of the presented algorithm is high, and measurement repeatability reaches 40 nm under the same experiment condition.展开更多
Advanced testing methods for the dynamics of mechanical microdevices are necessary to develop reliable, marketable microelectromechanical systems. A system for measuring the nanometer motions of microscopic structures...Advanced testing methods for the dynamics of mechanical microdevices are necessary to develop reliable, marketable microelectromechanical systems. A system for measuring the nanometer motions of microscopic structures has been demonstrated. Stop-action images of a target have been obtained with computer microvision, microscopic interferometry, and stroboscopic illuminator. It can be developed for measuring the in-plane-rigid-body motions, surface shapes, out-of-plane motions and deformations of microstructures. A new algorithm of sub-pixel step length correlation template matching is proposed to extract the in-plane displacement from vision images. Hariharan five-step phase-shift interferometry algorithm and unwrapping algorithms are adopted to measure the out-of-plane motions. It is demonstrated that the system can measure the motions of solder wetting in surface mount technology(SMT).展开更多
On account of the multiformity of MEMS devices, it is necessary to integrate with some optical measurement techniques for meeting static and dynamic unit test requirements. In this paper, an automated MEMS test system...On account of the multiformity of MEMS devices, it is necessary to integrate with some optical measurement techniques for meeting static and dynamic unit test requirements. In this paper, an automated MEMS test system is built of some commercially available components and instruments based on virtual instrument technology. The system is integrated with stroboscopic imaging, computer micro-vision, microscopic Mirau phase shifting interferometry, and laser Doppler vibrometer, and is used for the measurement of full-view in-plane and out-of-plane geometric parameters and periodical motions and single spot out-of-plane transient motion. The system configuration and measurement methods are analyzed, and some applications of the measurement of in-plane and out-of-plane dimensions and motions were described. The measurement accuracy of in-plane dimensions and out-of-plane dimensional is better than 0.2 um and 5 nm respectively. The resolution of measuring in-plane and out-of-plane motions is better than 15 nm and 2 nm respectively.展开更多
基金Supported by Youth Natural Science Foundation of Beijing University of Chemical Technology (No.QN0734).
文摘Optical flow method is one of the most important methods of analyzing motion images. Optical flow field is used to analyze characteristics of motion objects. According to motion features of micro-electronic mechani-cal system (MEMS) micro-structure, the optical algorithm based on label field and neighborhood optimization is presented to analyze the in-plane micro-motion of micro-structure. Firstly, high speed motion states for each fre-quency segment of micro-structure in cyclic motion are frozen based on stroboscopic principle. Thus a series of dynamic images of micro-structure are obtained. Secondly, the presented optical algorithm is used to analyze the image sequences, and can obtain reliable and precise optical field and reduce computing time. As micro-resonator of testing object, the phase-amplitude curve of micro-structure is derived. Experimental results indicate that the meas-urement precision of the presented algorithm is high, and measurement repeatability reaches 40 nm under the same experiment condition.
文摘Advanced testing methods for the dynamics of mechanical microdevices are necessary to develop reliable, marketable microelectromechanical systems. A system for measuring the nanometer motions of microscopic structures has been demonstrated. Stop-action images of a target have been obtained with computer microvision, microscopic interferometry, and stroboscopic illuminator. It can be developed for measuring the in-plane-rigid-body motions, surface shapes, out-of-plane motions and deformations of microstructures. A new algorithm of sub-pixel step length correlation template matching is proposed to extract the in-plane displacement from vision images. Hariharan five-step phase-shift interferometry algorithm and unwrapping algorithms are adopted to measure the out-of-plane motions. It is demonstrated that the system can measure the motions of solder wetting in surface mount technology(SMT).
基金Supported by National Natural Science Foundation of China (No.50505031)Tianjin Municipal Science and Technology Commission(No.043185911)Programfor NewCentury Excellent Talents in University
文摘On account of the multiformity of MEMS devices, it is necessary to integrate with some optical measurement techniques for meeting static and dynamic unit test requirements. In this paper, an automated MEMS test system is built of some commercially available components and instruments based on virtual instrument technology. The system is integrated with stroboscopic imaging, computer micro-vision, microscopic Mirau phase shifting interferometry, and laser Doppler vibrometer, and is used for the measurement of full-view in-plane and out-of-plane geometric parameters and periodical motions and single spot out-of-plane transient motion. The system configuration and measurement methods are analyzed, and some applications of the measurement of in-plane and out-of-plane dimensions and motions were described. The measurement accuracy of in-plane dimensions and out-of-plane dimensional is better than 0.2 um and 5 nm respectively. The resolution of measuring in-plane and out-of-plane motions is better than 15 nm and 2 nm respectively.