In insect flapping wings, the camber deformation is caused by the aerodynamic forces. Since the camber will improve the aerodynamic performance of Flapping Wing Nano Air Vehicles (FWNAVs), it is important to elucidate the passive mechanism of the cambering. The pixel wing model consisting of a structured mesh using shell elements that can simulate the camber deformation caused by the fluid–structure interaction has been proposed for the purpose of computational efficiency. In this study, the performance of the pixel wing model is evaluated as the pixel model resolution is changed. The minimum pixel model resolution is determined such that it can keep enough magnitude of camber compared to actual insects. Furthermore, it is found that the cambering of the pixel wing model can be effectively changed using the wing's chord-wise flexural stiffness given by the root vein pixels and the thickness of the wing membrane pixels.

The number of double-strand breaks can be evaluated from the change of average DNA length. The average DNA length is measured by the single-molecule observation method using fluorescence microscope. The measurement of DNA length in the microscope images is done manually by experienced operators and it is time consuming in many experiments. An image processing method using OpenCV library to measure length of DNA in fluorescence microscope images is developed in this paper. An automation of measurement using deep learning is also proposed.

The purpose of this study is an improvement of convergence properties of an interface problem in the high-frequency electromagnetic field finite element analysis based on the iterative domain decomposition method by introducing a pseudo-quadruple precision into the subdomain solver. In the high-frequency electromagnetic field analysis, it is well known that when the numerical model becomes huge scale, the convergence properties of the interface problem become extremely poor. To improve the deterioration of the convergence property in the interface problem, we propose to introduce a pseudo-quadruple precision into the iterative solver based on the conjugate gradient method in the subdomain problem. The pseudo-quadruple precision is constructed by ordinary double precisions. In this study, we apply the pseudo-quadruple precision for solving subdomain problems in the iterative domain decomposition method. As a result, we can obtain improvement of convergence properties in the iterative solver for the interface problem.

Molecular dynamics simulations of the hydrogen-removed polyethylene are carried out to study the structural change of polyethylene induced by beta decays of substituted tritium. Our simulations show that the folded structure of the hydrogen-removed polyethylene becomes more disordered as the number of removed hydrogen atoms becomes larger. We also propose a theoretical approach to explaining and predicting our molecular dynamics simulation results of hydrogen-removed polyethylene on the basis of the linear response theory. We derive the time derivative of the dynamical quantity, which is conjugate to the force applied as perturbation in the framework of the linear response theory, required to calculate the response function. The dynamical quantity in this study is the total potential energy difference of polyethylene before and after removal of hydrogen. Preliminary results of the response function for the total potential energy of polyethylene after removal of hydrogen are presented.

Our work aims to interpret historical documents in closed booklet form without physical unfolding through CT scanning techniques. X-ray based CT scans can transform booklets into physical data for analysis, and current research in this direction performs page segmentation at the two-dimensional level and does not analyze data from historical documents at the three-dimensional level. Our proposed method uses a self-developed web tool for 3D dimensional annotation and solves Laplace’s equation to construct the 3D spatial structure of historical documents and perform page segmentation. We experiment with data in booklet form and show how to segment and extract pages in three dimensions by using Laplace’s equation.