Metallography is the study of the structure of metals and alloys. Metallographic analysis can be regarded as a detection tool to assist in identifying a metal or alloy, to evaluate whether an alloy is processed correctly, to inspect multiple phases within a material, to locate and characterize imperfections such as voids or impurities, or to find the damaged areas of metallographic images. However, the defect detection of metallography is evaluated by human experts, and its automatic identification is still a challenge in almost every real solution. Deep learning has been applied to different problems in computer vision since the proposal of AlexNet in 2012. In this study, we propose a novel convolutional neural network architecture for metallographic analysis based on a modified residual neural network (ResNet). Multi-scale ResNet (M-ResNet), the modified method, improves efficiency by utilizing multi-scale operations for the accurate detection of objects of various sizes, especially small objects. The experimental results show that the proposed method yields an accuracy of 85.7% (mAP) in recognition performance, which is higher than existing methods. As a consequence, we propose a novel system for automatic defect detection as an application for metallographic analysis.

Due to high demand for the safe operation of structures, structural members with the potential to shift their natural frequencies/specific-stiffness to satisfy design requirements or avoid the occurrence of dynamical instability are of great importance. Hence, the purpose of this study is to present a methodology to evaluate the impact arising from the number of layer and properties of laminas on the natural frequencies of a passive periodic-array beam. The periodic array is assumed to be composed of two different materials, basic and embedded laminas, which are periodically embedded into the beam. The problem is solved by treating the periodic-array beam as a single-phase nonhomogeneous continuum. The variation of material properties in the axial direction is expressed by a Fourier series half-range expansion with a wavelength equal to the average space between the two materials. The Fourier-series-based approach and Newtonian mechanics are employed in the analysis. The result of the study indica...