Demands for machining of precision parts have recently increased. In order to meet such requirements, not only ultra-precision machining systems, but also precision profile measuring systems are required. In the profile measurement of precision parts, since the measurement marks have a large influence, a completely non-contact shape measurement method free from measurement marks is required. In the profile measurement using the optical probe, the profile of a specimen is measured by keeping the distance between a lens and a surface constant while scanning. Since the importance of measuring the complex three-dimensional shape increases, the importance of an optical probe that can measure such a shape in a non-contact condition increases. In this paper, problems and research examples in profile measurement using optical probes with high measuring accuracy are explained.

　　　We had reported a 3D-shape measurement technique of a rough surface using speckle interferograms captured under random phase shift. Amounts of the phase shifts between captured interferograms are calculated both through Max & Min light intensities searchings at two different pixels along frames and twin normalizations of their intensity changes. The calculated phase shifts are used to extract phase distribution of a speckle interferogram. After capturing 280 specklegrams which consist of 40 specklegrams at wavelengths of 778.19 nm, 778.20 nm, 778.24 nm 778.35 nm, 778.61 nm, 779.14 nm and 780.09 nm, phase distributions are extracted regarding the first captured interferogram in each wavelength. The ratios of phase change against wavelength change are then extracted at all the pixels, and 3D shape is calculated from the extracted ratios. This 3D-shape measurement technique is now applied to an iron head of electric-solder whose temperature is rising from room temperature, RT, to 392 degree C. At the first trial, the obtained shape data have involved intense error except RT. Then we introduce some new data processing, and then analyze origins of remaining errors. Finally, effective data processing are found which can measure the 3D shape of the head at from RT to 390 degree C.

　　　Bearing part is one of the most vulnerable parts of railway bridges. The soundness of railway bridges is usually inspected visually. However, visual inspection to detect deterioration at bearing part is sometimes difficult when trains run on railway bridges. Therefore, deflection angle at bearing part has been used to evaluate the soundness of bearing part of railway bridges. Sampling moire camera is one of the measuring equipment of deflection angle. It can measure deflection angle in safe and without contact. However, a measured value of deflection angle includes a lot of noise due to vibration of substructures, distance between the sampling moire camera and bearing parts, etc. In this paper, improvement in the precision of deflection angle measured by the sampling moire camera was investigated. In addition, practical applicability of the soundness evaluation of bearing part by measuring deflection angle was verified. The results show that deflection angle calculated from a difference of vertical displacement between two targets has enough precision to evaluate the soundness of bearing part of railway bridges.

　　　Measurement of speed has a very important role in various industrial and academic fields. In general, the speed measurement is limited to either the measurement of the relative speed with respect to the reference point or the integration of the acceleration. In the measurement method using the reference point, the reference point needs to be immovable, so it cannot be used for vibration measurement or the like. On the other hand, the method of integrating acceleration has a problem that errors accumulate. Angular velocity can be measured directly by using optical gyroscope, and it is used for attitude control of aircraft. In this study, a direct measurement method of straight travelling speed by applying the principle of optical gyroscope was proposed. As a result of the verification experiment of the measurement principle, correlation between the linear speed and the frequency change of interference light output was obtained, and the possibility of measuring the linear speed by the proposed method was shown.

　　　Many underground storage tanks are used for decades after they are installed. Therefore, rust occurs on the walls of these tanks as a result of local corrosion. Leaks from the tanks occur as the corrosion progresses. Such tanks are monitored and checked through pressure tests or observation tests. However, the corrosion cannot be detected in the absence of tank leakage. The acoustic emission (AE) method is employed to monitor the corrosion by detecting fractures in the rust. In this study, we used the finite-difference time-domain method in a simulation to determine the ideal sensor shape for an underground storage tank. First, a method was determined for placing the AE sensor in a rectangular solid case to effectively detect the AE using a model with a fluid solid boundary, assuming an underground storage tank. The simulation results indicated that the amplitude of the wave propagating in the upper and lower portions of the tank was large. AE signals from the bottom of the tank were effectively detected by the sensor mounted on the bottom of the sensor box. AE signals from the side wall were effectively detected on the upper part of the side wall of the sensor box. Next, immersion type AE sensor was constructed by using the result of simulation and sensitivity was studied. Amplitude distribution of developed system from different artificial AE source was corresponding to the simulation result. Sensitivity of the developed sensor was compared to the immersion sensor with single optical fiber sensor. Developed system had higher sensitivity than the single optical fiber sensor system.

　　　Effects of relative positions of the random pattern and the imaging sensor for measurement error in digital image correlation method are investigated. In order to take account of the effect of the relative position of microlens array on the imaging sensor and the speckle image formed on the image plane, simulations of imaging processes are carried out. Errors of DIC measurement are evaluated using the simulated images having several displacements and initial gap between the center of the pixel and the speckle. As a result, relationships between the error and the displacements for each center gap show various profiles. The average profile of them shows a sinusoidal curve with a relatively low amplitude. Since various error profiles of each speckle in the random pattern are averaged in the subset area in DIC measurement of translation displacements, the error profile of the subset approaches the average sinusoidal curve.

　　　Accurate 3D shape measurement systems by contactless method are required in the quality inspections electronic parts in industrial fields. In this paper, a multi-line fiber array with LED light sources for 3D shape measurement is prototyped. In a previous research, authors developed a measurement system with Talbot fringe projection using super luminescent diodes (SLD). However, the occurrence of speckle noise due to coherency in SLD is a factor to increase the noise of projected fringe pattern. In the method of shape measurement by analysis of grating images, the speckle noise gives large influences to the results. A light source with enough narrower width than the grating pitch is required to generate a Talbot fringe pattern. Therefore, authors developed a multi-line bundle fibers with LED light sources. The effectiveness to reduce the speckle noise was evaluated as an experimental result.

　　　This paper presents a method for identifying dynamic characteristics of a bridge using stochastic system identification and statistical analysis. In this study, three-dimensional shape of a truss bridge is obtained by 3D laser scanning and used for FE modeling. The initial FE model based on 3D measurement is validated and updated by comparing analytical dynamic characteristics of the FE model and the actual characteristics of the bridge estimated by stochastic system identification. Estimations of dynamic characteristics generally fluctuate under the influence of the environment in service state. In this study, distribution of the series of estimated frequencies is assumed as a Kernel distribution and significant frequencies such as natural frequencies are classified based on the probability density of estimations. The method stably provides accurate characteristics of the bridge under several excitation states.

　　　It is important both the design and the mechanical function of the tread pattern (groove) of automobile tires which referred to as the face of the tire. Currently, it is required to achieve of improvement from the viewpoint of both its mechanical function and design. In this study, in order to clarify the mechanical function of tire, the effect of the tread pattern on the strain, the maximum static friction force and the actual contact ratio is evaluated. Moreover, the relationship between the evaluated parameters and the maximum static friction force are discussed for considering the mechanical function of the tire. This paper describes not only the results of simultaneous measurement of strain and maximum static friction force by DIC method but also the measurement result of actual contact ratio using the transfer method.