This study proposes a lightweight convolutional neural network with an octave-like convolution mixed block, called OLCMNet, for detecting driver distraction under a limited computational budget. The OLCM block uses point-wise convolution (PC) to expand feature maps into two sets of branches. In the low-frequency branches, we perform average pooling, depth-wise convolution (DC), and upsampling to obtain a low-resolution low-frequency feature map, reducing spatial redundancy and connection density. In the high-frequency branches, the expanded feature map with the original resolution is fed to the DC operator, gaining an apposite receptive field to capture fine details. The feature concatenation of the low-frequency and high-frequency branches is encoded sequentially by a squeeze-and-excitation (SE) module and PC operator, realizing feature global information fusion. Introducing another SE module at the last stage, the OLCMNet facilitates further sensitive information exchange between layers. In addition, with an augmented reality head-up display (ARHUD) platform, we create a Lilong Distracted Driving Behavior (LDDB) Dataset through a series of on-road experiments. Such a dataset contains 14808 videos collected from an infrared camera, covering six driving behaviors of 2468 participants. We manually annotate these videos at five frames per second, obtaining a total of 267378 images. Compared with the existing methods, the embedded hardware platform experiments indicate that OLCMNet hits acceptable trade-offs, namely, 89.53% accuracy for StateFarm Dataset and 95.98% accuracy LDDB Dataset when the latency is 32.8 ± 4.6ms.

To improve the wear monitoring accuracy of steel tribo-pairs, a ferromagnetic particle sensor that consists of a magnetic loop with two cylindrical magnetic poles and one magnetic housing is proposed. The induction coil is placed inside the magnetic poles to shield against external electromagnetic interference. The air gap is designed to 1 mm between the two magnetic poles, where a rotational symmetric high-gradient static magnetic field will be generated under constant direct current (dc) drive. The sine-wave-like signal of one period will be induced when one ferromagnetic particle flows through the sensor. The working principle of the sensor and the internal magnetic field distribution is explained by establishing a mathematical model and finite element analysis (FEA). Furthermore, the results about the effects of coil length and position show that the output voltage is not sensitive to the axial mounting positions of the induction coil, and a shorter inductive coil produces a larger output voltage. By the oil dynamic monitoring experiment, the sensor can distinguish 13- $\mu \text{m}$ ferromagnetic particles in an oil tube of 10-mm inner diameter. With its high sensitivity to ferromagnetic particles, the sensor is suitable for online wear monitoring of steel tribo-pairs.

With the rapid development of wearable smart electronic products, high-performance wearable flexible strain sensors are urgently needed. In this paper, a flexible strain sensor device with Fe NWs/Graphene/PEDOT:PSS material added under a porous structure was designed and prepared. The effects of adding different sensing materials and a different number of dips with PEDOT:PSS on the device performance were investigated. The experiments show that the flexible strain sensor obtained by using Fe NWs, graphene, and PEDOT:PSS composite is dipped in polyurethane foam once and vacuum dried in turn with a local linearity of 98.8%, and the device was stable up to 3500 times at 80% strain. The high linearity and good stability are based on the three-dimensional network structure of polyurethane foam, combined with the excellent electrical conductivity of Fe NWs, the bridging and passivation effects of graphene, and the stabilization effect of PEDOT:PSS, which force the graphene-coated Fe NWs to adhere to the porous skeleton under the action of PEDOT:PSS to form a stable three-dimensional conductive network. Flexible strain sensor devices can be applied to smart robots and other fields and show broad application prospects in intelligent wearable devices.

In this paper, an all-metal metamaterial absorber as a sensitive biosensor is proposed in the terahertz regime for biomolecular detection, such as protein, bacillus bacteria, cancer cells, and so on. Here, the highly sensitive characteristic of the proposed absorber is implemented by the enhanced electromagnetic loss of symmetry-broken metal structures, where two discontinuous hollow rectangular metal arms are arranged orthogonally above the metal ground plane in geometry. Finite integral theory (FIT) based software is utilized for modeling and full-wave numerical simulation. The results show that a near-perfect absorption of 100% at 1.99 THz is achieved with a quality factor value of 87, which could be an excellent performance in the terahertz regime below 2 THz. The electric and magnetic field distribution analysis is performed to reveal the mechanism, that is the significant confinement of electromagnetic field in the specially designed metal structure. The sensing performance of a refractive index is also investigated, and the resonance frequency and peak absorption of the spectra will get shift due to the variation of biological samples. The maximum sensitivity and figure of merit reach 540 GHz/RIU and 23.5 RIU ^-1 , respectively. In addition, the effect of polarization angles and performance under different incident angles are also discussed. The presented biosensor with high sensitivity and feasible fabrication is a good candidate for biomedical detection in the terahertz regime.