In the single-pixel particle image velocimetry (PIV) method, the idea of the conventional method to calculate the variation is introduced and the extension of this method to the case in which the ideal particle image is not obtained is addressed. The extended method is based on the previous method which assumes the correlation distribution as a probability distribution, and it estimates the fluctuation by compensating the component corresponding to the shape of the particle using the Fourier transform. The average velocity field and the fluctuation field with very high resolution can be obtained for supersonic PIV by using the extended method. In the future, we expect such a proposal of a new technique for visualizing fluid phenomena with higher resolution using similar information processing.

A time-resolved phase microscopy has been developed for the observation of the photo-excited charge carrier dynamics in solar-devices such as photocatalysts and photovoltaics. By utilizing the pump-probe technique for the imaging, the time-resolution is only limited by the pulse width, 5 ns. An arbitrary spatial pattern of a pump pulse light excites the charge carriers, which provides various image recovery techniques for the refractive index image with a low S/N ratio. We applied the image recovery using the robust principal component analysis and the sparsity in the Fourier domain. By utilizing the data assimilation with the charge decay model, the lifetime and diffusion coefficients were mapped out for the photo-excited electrons in inhomogeneous titanium oxide film.

In order to extract physicochemical processes related to rock formation, it is necessary to acquire quantitative data at the maximum extent from the rock samples by measurement and analysis, and to obtain essential information from the acquired data through quantitative modeling. This paper introduces the author’s researches using the integration approach between informatics and measurement. Two typical petrological problem settings are discussed: process extraction from micro texture in a single rock sample and process extraction from bulk chemical compositions of multiple rock samples. In both of problem settings, visualization plays an important role in promoting the utilization of the expertise of a petrologist in a series of research processes such as acquisition of measurement data, interpretation of processes, and quantitative modeling. In the future, it is especially desirable to develop novel visualization and information-processing methods that promote the discovery of unknown rock-formation processes by considering the characteristics of geoscientific data.

Through the images on the retina, we obtain the visual qualities called as ‘Shitsukan’ in which they appear to be shiny, diffuse, etc. ‘Shitsukan’ is strongly related to the reflective properties. The reflection property is the ratio of the observed light to the incident light through the object. This characteristic varies depending on the direction of light incidence and the direction of observation. Because of the multidimensional nature of the reflective properties, it takes time to measure them. However, if we can speed up the measurement, it is highly likely to be useful in a wide range of application fields such as digital archiving, production, inspection, video production, and so on. Based on the background, this paper aims at minimizing the number of sampling in the measurements of the reflective properties. This paper introduces a method that can achieve a high-speed and high-accurate performance by fusing the information and measurement technologies.

  This paper introduces the bi-directional processing method using both radar and tomography imaging methods, assuming microwave non-destructive testing for concrete material. In particular, we introduce highly accurate radar imaging method as range points migration (RPM) method to limit the region of interest (ROI), which contributes to a remarkable reduction of the number of unknowns in the tomographic approach. Furthermore, the ROI updating scheme by exploiting the cost function of the contrast source inversion (CSI) method is presented to achieve a bi-directional effects between radar and tomography. The numerical simulation results demonstrated that our method significantly upgrades the accuracy for complex permittivity reconstruction and the ROI estimates.

X-ray micro-imaging is used to observe structures of materials and biological samples with high resolution by taking advantage of the short wavelength and high transmission of X-rays. In this study, we developed a high-resolution micro-imaging technique for liquid samples using a single femtosecond X-ray free-electron laser pulse. In the measurement using a focused X-ray free-electron laser, not only a sample but also sample holder itself is destroyed by high-intensity X-ray irradiation. Therefore, we developed a new high-integration solution sample holder for coherent diffraction imaging to realize efficient measurement of samples in liquid by X-ray free-electron laser. We succeeded in high-resolution imaging of nanoparticle structures in liquid using the newly developed solution sample holder at SACLA X-ray free-electron laser facility in Japan.

The question of how the organ-specific morphology is determined is one of the unsolved important issues in biology. Sor far, molecular biological studies have identified key genes for organ morphogenesis, but many of the physical processes on cell and tissue dynamics remain unknown. One of the reasons is that high-resolution measurement of cell/tissue dynamics is impossible even with the most advanced microscopes, except for limited tissues and species. In this paper, we introduce a Bayesian model for reconstructing deformation maps of curved epithelial tissues from low-resolution data on cell trajectories during organ development. After explaining the basic idea of modeling, we apply this method to the measurement data of chick early forebrain development, and show the validity of the method.

Total-scattering measurements using high-brilliance synchrotron X-rays allow one to visualize heterogeneous phenomena inside a bulk on the atomic scale. However, such measurements have not been achieved yet because of noises caused by measurements and analysis. This paper reports a data-driven approach to correct X-ray response non-uniformity, which has been recognized as one of the problems in X-ray detectors.