Dissimilar joining of aluminum and steel is one of the fundamental techniques for achieving weight reduction in the automotive industry. Self-piercing rivets have been widely adopted for the aluminum-steel lap joint because of its simplicity and high joining strength. In this study, the joining quality of the aluminum-steel joint produced by self-piercing rivets is investigated depending on the strength and thickness of the aluminum sheet. Four different tensile strengths (228, 305, 320, and 326 MPa) of aluminum are considered in this study, which corresponds to Al5052-H32, Al5754-H18, Al5083-H32 and Al6061-T6. For the thickness variation test, Al5052-H32 sheets with a thickness of 1.2, 1.5, 2.0, 2.5 mm are used. The SPR joint quality was quantified with cross-sectional visual analysis and tensile lap shear test. In aluminum(top)-steel (bottom) SPR joint, it was observed that the joint load increased with increasing aluminum strength up to 320 MPa. Nevertheless, the thickness of top aluminum was found to not affect the joint strength between the thickness of 1.5 ~ 2.5 mm. In steel(top)-aluminum(bottom) SPR joint, it was found that joint strength did not change with aluminum strength but increased with increasing aluminum thickness.

Medical image analyses have been widely used to differentiate normal and abnormal cases, detect lesions, segment organs, etc. Recently, owing to many breakthroughs in artificial intelligence techniques, medical image analyses based on deep learning have been actively studied. However, sufficient medical data are difficult to obtain, and data imbalance between classes hinder the improvement of deep learning performance. To resolve these issues, various studies have been performed, and data augmentation has been found to be a solution. In this review, we introduce data augmentation techniques, including image processing, such as rotation, shift, and intensity variation methods, generative adversarial network-based method, and image property mixing methods. Subsequently, we examine various deep learning studies based on data augmentation techniques. Finally, we discuss the necessity and future directions of data augmentation.

No abstract