Wi-AM: Enabling Cross-Domain Gesture Recognition with Commodity Wi-Fi.

RF-based gesture recognition systems outperform computer vision-based systems in terms of user privacy. The integration of Wi-Fi sensing and deep learning has opened new application areas for intelligent multimedia technology. Although promising, existing systems have multiple limitations: (1) they only work well in a fixed domain; (2) when working in a new domain, they require the recollection of a large amount of data. These limitations either lead to a subpar cross-domain performance or require a huge amount of human effort, impeding their widespread adoption in practical scenarios. We propose Wi-AM, a privacy-preserving gesture recognition framework, to address the above limitations. Wi-AM can accurately recognize gestures in a new domain with only one sample. To remove irrelevant disturbances induced by interfering domain factors, we design a multi-domain adversarial scheme to reduce the differences in data distribution between different domains and extract the maximum amount of transferable features related to gestures. Moreover, to quickly adapt to an unseen domain with only a few samples, Wi-AM adopts a meta-learning framework to fine-tune the trained model into a new domain with a one-sample-per-gesture manner while achieving an accurate cross-domain performance. Extensive experiments in a real-world dataset demonstrate that Wi-AM can recognize gestures in an unseen domain with average accuracy of 82.13% and 86.76% for 1 and 3 data samples.

The characterization of bovine compact bone fatigue damage using terahertz spectroscopy.

Fatigue can cause cracks to propagate from the micro- to the macroscale, which results in a decrease of Young's modulus of the bone. Non-destructive measurements of bone fatigue damage are of great importance for bone quality assessment and fracture prevention. Unfortunately, there is still a lack of effective nondestructive methods sensitive to the initial deterioration during damage accumulation, particularly in the field of orthopedics and biomechanics. In this study, terahertz spectroscopy was adopted to evaluate microscale bone damage. Specifically, the refractive index and Young's modulus of bone samples subjected to different degrees of fatigue damage were tested at a fixed area. Both parameters are found to decrease in two stages under cycled fatigue loading, which is attributed to the initial onset and subsequent development of microdamage during fatigue loading. The change in refractive index reflects the accumulation of fatigue damage as well as the decrease in Young's modulus.

Modeling of Bimodular Bone Specimen under Four-Point Bending Fatigue Loading.

The fatigue damage behavior of bone has attracted significant attention in both the mechanical and orthopedic fields. However, due to the complex and hierarchical structure of bone, describing the damage process quantitively or qualitatively is still a significant challenge for researchers in this area. In this study, a nonlinear bi-modulus gradient model was proposed to quantify the neutral axis skewing under fatigue load in a four-point bending test. The digital image correlation technique was used to analyze the tensile and compressive strains during the fatigue process. The results showed that the compressive strain demonstrated an obvious two-stage ascending behavior, whereas the tensile strain revealed a slow upward progression during the fatigue process. Subsequently, a theoretical model was proposed to describe the degradation process of the elastic modulus and the movement of the neutral axis. The changes in the bone properties were determined using the FEM method based on the newly developed model. The results obtained from two different methods exhibited a good degree of consistency. The results obtained in this study are of help in terms of effectively exploring the damage evolution of the bone materials.

Tensile and compressive strain evolutions of bovine compact bone under four-point bending fatigue loading.

Bones are biological composite materials with multiscale structures. Bone fatigue damage is commonly characterized by an increase in strain that is accompanied by microdamage at different scales. This study investigated the damage evolutions of bone specimens under four-point bending fatigue loading using neutral axis migration. Tensile and compressive strains during the fatigue process were simultaneously measured using a digital image correlation technique. The compressive strain of the bone specimen increased rapidly at first and then proceeded slowly while the tensile strain decreased during fatigue loading. Consequently, the neutral axis shifted downward as the damage accumulated. A positive correlation exists between the downward offset of the neutral axis and the number of cycles. The variation in compressive strain is larger than that in tensile strain in this situation.

Effects of Fatigue Damage on the Microscopic Modulus of Cortical Bone Using Nanoindentation.

Alterations to the bone structure from cycle loadings can undermine its damage resistance at multiple scales. The accumulation of fatigue damage in a bone is commonly characterized by the reduction in the elastic modulus. In this study, nano-indentation was used for investigating microscopic damage evolution of bovine tibia samples subjected to fatigue loading. Indentation tests were conducted in the same 60 µm × 120 µm area with different degrees of damage, including fracture, and the evolution of reduced modulus was observed. The results showed that bone's reduced modulus decreased significantly during the initial 40% of the life fraction, whereas it proceeded slowly during the remaining period. As the size of the residual indentations was about 4 µm in length, the degradation of bone's reduced modulus reflected the accumulation of fatigue damage at smaller scales.