In the intelligent traffic system, real-time and accurate detections of vehicles in images and video data are very important and challenging work. Especially in situations with complex scenes, different models, and high density, it is difficult to accurately locate and classify these vehicles during traffic flows. Therefore, we propose a single-stage deep neural network YOLOv3-DL, which is based on the Tensorflow framework to improve this problem. The network structure is optimized by introducing the idea of spatial pyramid pooling, then the loss function is redefined, and a weight regularization method is introduced, for that, the real-time detections and statistics of traffic flows can be implemented effectively. The optimization algorithm we use is the DL-CAR data set for end-to-end network training and experiments with data sets under different scenarios and weathers. The analyses of experimental data show that the optimized algorithm can improve the vehicles’ detection accuracy on the test set by 3.86%. Experiments on test sets in different environments have improved the detection accuracy rate by 4.53%, indicating that the algorithm has high robustness. At the same time, the detection accuracy and speed of the investigated algorithm are higher than other algorithms, indicating that the algorithm has higher detection performance.

In this study, Nb₂O₅ ceramic was used as the target to deposit the Nb₂O₅ thin films on glass substrates with the radio frequency (RF) magnetron sputtering method. Different deposition temperatures and O₂ ratios were used as parameters to investigate the optical properties of Nb₂O₅ thin films. The deposition parameters were a pressure of 5 × 10⁻³ Torr, a deposition power of 100 W, a deposition time of 30 min, an O₂ ratio (O₂/(O₂ + Ar), in sccm) of 10% and 20%, and deposition temperatures of room temperature (RT), 200, 300 and 400 °C, respectively. We found that even if the deposition temperature was 400 °C, the deposited Nb₂O₅ thin films revealed an amorphous phase and no crystallization phase was observed. The optical properties of transmittance of Nb₂O₅ thin films deposited on glass substrates were determined by using a ultraviolet-visible (UV-vis) spectrophotometer (transmittance) and reflectance spectra transmittance (reflectance, refractive index, and extinction coefficient) in the light wavelength range of 250–1000 nm. When the O₂ ratio was 10% and the deposition temperature increased from RT to 200 °C, the red-shift was clearly observed in the transmittance curve and the transmission ratio had no apparent change with the increasing deposition temperature. When the O₂ ratio was 20%, the red-shift was not observed in the transmittance curve and the transmission ratio apparently decreased with the increasing deposition temperature. The variations in the optical band gap (E_g) values of Nb₂O₅ thin films were evaluated from the Tauc plot by using the quantity hν (the photon energy) on the abscissa and the quantity (αhν)^r on the ordinate, where α is the optical absorption coefficient, c is the constant for direct transition, h is Planck’s constant, ν is the frequency of the incident photon, and the exponent r denotes the nature of the transition. As the O₂ ratio of 10% or 20% was used as the deposition atmosphere, the measured E_g values decreased with the increase of the deposition temperature. The reflectance ratio, extinction coefficient, and refractive index curves of Nb₂O₅ thin films were also investigated in this study. We would show that those results were influenced by the deposition temperature and O₂ ratio.

We theoretically proposed and numerically analyzed a polarization-independent, wide-angle, and ultra-broadband absorber based on a multi-layer metasurface. The numerical simulation results showed that the average absorption rates were more than 97.2% covering the broad wavelength of 400~6000 nm (from visible light to mid-infrared light) and an absorption peak was 99.99%, whatever the polarization angle was changed from 0° to 90°. Also, as the incidence angle was swept from 0° to 55°, the absorption performance had no apparent change over the wavelength ranges of 400 to 6000 nm. We proved that the proposed metasurface structure was obviously advantageous to achieve impedance matching between the absorber and the free space as compared with conventionally continuous planar-film structures. The broadband and high absorption resulted from the strong localized surface plasmon resonance and superposition of resonant frequencies. As expectable the proposed absorber structure will hold great potential in plasmonic light harvesting, photodetector applications, thermal emitters and infrared cloaking.

We investigate a high-sensitivity surface plasmon resonance (SPR) biosensor consisting of a Au layer, four-layer MoS₂, and monolayer graphene. The numerical simulations, by the transfer matrix method (TMM), demonstrate the sensor has a maximum sensitivity of 282°/RIU, which is approximately 2 times greater than the conventional Au-based SPR sensor. The finite difference time domain (FDTD) indicates that the presence of MoS₂ film generates a strong surface electric field and enhances the sensitivity of the proposed SPR sensor. In addition, the influence of the number of MoS₂ layers on the sensitivity of the proposed sensor is investigated by simulations and experiments. In the experiment, MoS₂ and graphene films are transferred on the Au-based substrate by the PMMA-based wet transfer method, and the fabricated samples are characterized by Raman spectroscopy. Furthermore, the fabricated sensors with the Kretschmann configuration are used to detect okadaic acid (OA). The okadaic acid–bovine serum albumin bioconjugate (OA-BSA) is immobilized on the graphene layer of the sensors to develop a competitive inhibition immunoassay. The results show that the sensor has a very low limit of detection (LOD) of 1.18 ng/mL for OA, which is about 22.6 times lower than that of a conventional Au biosensor. We believe that such a high-sensitivity SPR biosensor has potential applications for clinical diagnosis and immunoassays.

We propose a high-performance algorithm while using a promoted and modified form of the You Only Look Once (YOLO) model, which is based on the TensorFlow framework, to enhance the real-time monitoring of traffic-flow problems by an intelligent transportation system. Real-time detection and traffic-flow statistics were realized by adjusting the network structure, optimizing the loss function, and introducing weight regularization. This model, which we call YOLO-UA, was initialized based on the weight of a YOLO model pre-trained while using the VOC2007 data set. The UA-CAR data set with complex weather conditions was used for training, and better model parameters were selected through tests and subsequent adjustments. The experimental results showed that, for different weather scenarios, the accuracy of the YOLO-UA was ~22% greater than that of the YOLO model before optimization, and the recall rate increased by about 21%. On both cloudy and sunny days, the accuracy, precision, and recall rate of the YOLO-UA model were more than 94% above the floating rate, which suggested that the precision and recall rate achieved a good balance. When used for video testing, the YOLO-UA model yielded traffic statistics with an accuracy of up to 100%; the time to count the vehicles in each frame was less than 30 ms and it was highly robust in response to changes in scenario and weather.

Here, we have theoretically proposed an ideal structure of selective solar absorber with multilayer planar films, which can absorb the incident light throughout the entire solar spectrum (300–2500 nm) and over a wide angular range, whatever the polarization angle of 0°~90°. The efficiency of the proposed absorber is proven by the Finite-Difference Time Domain (FDTD) simulation. The average absorption rate over the solar spectrum is up to 96.6%. The planar design is extremely easy to fabricate and modify, and this structure does not require lithographic processes to finish the absorbers. Improvements of the solar absorber on the basis of planar multilayer-film structures is attributed to multiple asymmetric highly lossy Fabry–Perot resonators. Because of having many virtues, such as using different refractory and non-noble metals, having angle and polarization independence, and having ideal absorption for entire solar spectrum, our proposed absorbers are promising candidates for practical industrial production of the solar-energy harvesting.