Traditional deep convolutional networks (ConvNets) have shown that both RGB and depth are complementary for video action recognition. However, it is difficult to enhance the action recognition accuracy because of the limitation of the single ConvNets to extract the underlying relationship and complementary features between these two kinds of modalities. In this paper, we proposed a novel two stream ConvNet for multi-modality action recognition by joint optimization learning to extract global features from RGB and depth sequences. Specifically, a non-local multi-modality compensation block (NL-MMCB) is introduced to learn the semantic fusion features for the recognition performance. Experimental results on two multi-modality human action datasets, including NTU RGB+D 120 and PKU-MMD dataset, verify the effectiveness of our proposed recognition framework and demonstrate that the proposed NL-MMCB can learn complementary features and enhance the recognition accuracy.
A hydrogen sensing transistor fabricated by a heterojunction bipolar transistor (HBT) with an extended base (EB) formed by a metal-semiconductor-metal (MSM) hydrogen sensor is reported. The power consumption in stand-by mode is smaller than 2 μW. Common-emitter characteristics show that the sensing base (collector) current gains at 25℃ in 0.01%, 0.1%, and 1% H2/N2 are as high as 75 (512), 134, (977), and 233 (2.89 104), respectively. Low-power consumption and high-sensitive gains are indicative that our HBT together with planar-type MSM sensor is very promising for applications to hydrogen sensing transistors using one voltage source.
Tetracalcium phosphate (TTCP) is one of the main powder components in self-setting calcium phosphate cements for hard tissue applications. In this study, two types of calcium phosphate/chondroitin sulfate bone cements in which TTCP powders in nanoscale-rod like (R-TTCP) and micro-conventional irregular shape (C-TTCP) were used. The first one was synthesized by reverse microemulsion chemical process and the second one, was prepared by thermal conventional method. The results showed that both cements formed hydroxyapatite as the result of cementation process. The R-TTCP cement revealed a slightly longer initial but no difference in final setting time, less compressive strength, higher porosity and better degradation behavior compared to C-TTCP one. The both cements presented similar tendency to the formation of a dense hydroxyapatite on their outer surfaces through immersion in simulated body fluid. Taking into consideration the initial porosity, the cement made from R-TTCP rod like nanopowder presented more aptness to participate in ion exchange in SBF resulting to fill the 15% more initial porosity via the precipitation of hydroxyapatite mineral. From the biological point of view, analysis of cytotoxicity and MG63 osteoblastic-cell behavior proved that the both cements had good viability and proper cell adhesion and activity.
Introduction Axial spondylarthritis (axSpA) is a chronic inflammatory disease and commonly results in pain and joint stiffness. Using remote technology, such as a computer vision-aided system, has the potential to monitor functional movement and posture. Methods The validity of the remote technology measurement of functional movement and posture were tested cross-sectionally and compared to a standard clinical measurement by a physiotherapist. The feasibility of remote implementation was tested in a home environment. In addition, a cost-benefit analysis was conducted. Results Thirty-one participants with axSpA (42% female, 54(SD 13) years old and 27.4(SD 5.3) kg/m2) and 31 participants without back pain (65% female, 36(SD 10) years old and 25.9(3.7) kg/m2). In the axSpA group, the validity of assessment on cervical rotation, lumbar flexion, lumbar side flexion, shoulder flexion, hip abduction, tragus-to-wall and thoracic kyphosis showed significant moderate to strong correlation; in the non-back pain group, the same measures showed significant correlation ranging from weak to strong. Conclusions Remote technology systems in rehabilitation have the potential to reduce health inequality and improve cost and time effectiveness for both patients and the health system. Additionally, results show that using this Computer Vision-aided system in a home environment is a safe method.
Rydberg-atom electrometers have the remarkable advantages of self-calibration and high sensitivity. Based on the classical electromagnetic theory, a localized electric field enhancement structure of a hybrid rectangular resonator is proposed to improve the sensitivity of quantum microwave measurement. It should be noted that the prototype of the hybrid rectangular resonator is fabricated and measured at 9.925 GHz. The results of full-wave simulations show that the uniform and high electric field enhancement in the TE101 fundamental mode is realized. The transient process of resonance is simultaneously simulated, and the time to settle steady state is given as about 104 ns. As indicated through experimental results that the structure can reach 24 dB (enhancement factor of 15.8). As a result, the method proposed in this study, based on atomic measurement capabilities, enables us to improve the measurement sensitivity further and promotes the practical development of quantum microwave measurement technology.
Affective video content analysis is an active topic in the field of affective computing. In general, affective video content can be depicted by feature vectors of multiple modalities, so it is important to effectively fuse information. In this work, a novel framework is designed to fuse information from multiple stages in a unified manner. In particular, a unified fusion layer is devised to combine output tensors from multiple stages of the proposed neural network. With the unified fusion layer, a bidirectional residual recurrent fusion block is devised to model the information of each modality. Moreover, the proposed method achieves state-of-the-art performances on two challenging datasets, i.e., the accuracy value on the VideoEmotion dataset is 55.8%, and the MSE values on the two domains of EIMT16 are 0.464 and 0.176 respectively. The code of UMFN is available at: https://github.com/yunyi9/UMFN.
Three-dimensional (3D) electromagnetic (EM) parametric modeling and implementation of miniaturized lumped-element power divider (PD) based on symmetrical configuration is investigated. The system design and schematic diagrams of PDs are firstly demonstrated, subsequently, the 3D EM field and field-circuit co-simulation models of PD utilizing computer simulation technology (CST) and advanced design system (ADS) tools are detailedly presented. Finally, one prototype is fabricated and measured. Simultaneously, the gain of the PD is -3.96 dB @ 100 MHz and the experimental curve is in excellent agreement with the simulation. As a conclusion, the novel design methodology applied in PD has important practical engineering value, which can be also applied in millimeter-wave (mmW) circuit.
This paper is concerned with the formation control problem for a class of large-scale mobile sensor networks. The dynamic of mobile sensors are modeled by class of semilinear parabolic system, which is a class of partial differential equation(PDE) and has rich geometric family. In this model, the communication topology of agents is a chain graph and fixed. Leader feedback laws which designed in a manner to the boundary control of semilinear parabolic system allow the mobile sensors stable deployment onto planar curves. By constructing appropriate Lyapunov functional and using linear matrix inequality, several sufficient criteria are derived ensuring the mobile sensor networks to be globally asymptotically stable at the equilibrium. A simulation example is provided to demonstrate the usefulness of the proposed formation control scheme.
An ultra-wideband (UWB) antenna with dual band-notched characteristics is proposed in this letter. The fabricated prototype of the proposed antenna has a compact size of 30×24mm which operates at 3GHz to 11GHz and attained dual notch-bands at 3.3-3.7GHz and 5.15-5.825GHz. The result shows that the proposed antenna is compact, easy to fabricate and provides good characteristics in radiation patterns and time-domain behaviors. The simulated and measured results displayed good agreement over the entire operating frequency band. The proposed antenna can be used for wideband frequency requirement systems like indoor positioning.
Event-based cameras are sensitive to brightness changes and can capture rich temporal information with very high temporal resolution, which has great potential for motion segmentation of moving objects. Under static background, events are only triggered by motion of objects, thereby moving objects can be easily segmented. However, in many real-world applications, events are also be triggered by the motion of camera or background and submerge the ones corresponding to moving objects. In this letter, we propose an event-based motion segmentation method to segment moving small objects in events obtained from the wild. First, motion estimation is performed to align the events triggered by the background. Then, candidate events corresponding to moving objects or moving backgrounds are detected. Finally, motion information is adopted to segment the events of moving small objects from the ones triggered by the background. In addition, we develop the first dataset for event-based motion segmentation of small objects, namely EMSS. Experimental results demonstrate the effectiveness of our method and show that our method can achieve robust motion segmentation of small moving objects in the wild.
In this letter, a joint weighted power detector (JWPD) based on maximum a posterior probability (MAP) criteria is proposed for Willie aiming at two-hop covert communication scenario, which is a near optimal detector. Instead of only supervising one single phase, Willie combines the observations of two phases to make joint decision in the proposed scheme. The proposed scheme achieves lower probability of detection error (PDE) than the existing single-phase-detector (SPD) scheme and adding-power-directly-detector (APDD) scheme due to sufficient utilization of the two-phases observations. Numerical results demonstrate the benefit of our proposed scheme.
This paper studies chatter stability of composite cutter bar milling system in rotating coordinate frame. Based on the structural dynamic equation and regenerative milling force model of composite cutter bar in rotating coordinate frame, the continuous distributed chatter analysis model of composite cutter bar milling system is established. The stability of milling system with a rotary symmetric dynamic cutter bar is predicted by using the semi-discrete time domain method. Influences including internal damping, external damping, symmetrical and asymmetric laminates on the stability of milling system are analyzed, and the results obtained in rotating and fixed coordinate frame are compared. It is shown that the results are consistent for symmetrical cutter bar either in the rotating coordinate frame or in the fixed coordinate frame. A new chatter instability zone appears at high rotating speeds due to material internal damping of the rotating composite cutter bar.
Due to its stochastic nature, wind energy imposes unprecedented challenges on the power grid, and a properly scheduled reserve is essential to accommodate wind power’s intermittency and volatility. Many power reserve scheduling studies have considered the uncertainties of the renewable energy integration but few address how different wind speed forecast techniques influence the scheduling of reserves in the congested transmission networks. In this paper, three forecasting techniques: artificial neural network, autoregressive integrated moving average, and probability distribution function-based model are adopted to forecast one day of wind speed at Taylor, TX in 2012. To evaluate the impacts of the forecast techniques on power reserve scheduling, a stochastic reserve optimization model was developed to ensure the delivery of reserve in the event of transmission congestion and ramping constraints. A modified RTS-96 test system was employed and the results claim that different forecast models significantly affect the amount of scheduled up and down reserves in a stochastic reserve optimization problem. The level of operating reserve that is induced by wind is not constant during all hours of the day. Dynamic up and down reserves will be needed with a large scale of wind farm integration.
To improve the diversity and performance of the Mayfly Algorithm (MA), this letter adopts the mutation strategies in the process of MA. The opposition-based learning (OBL) and Cauchy mutation strategies are used to mutate the global optimal solution, and the artificial mutation operator is used in the offspring population. The hybrid mutation strategies are used in a cascaded structure. The performance of the proposed algorithms is demonstrated in simulations comparatively.
This paper highlights the impact of curved and flat vehicular plastic parts on the radiation characteristics of two dual-band antennas for C-V2X applications. The radiation patterns of the antennas are measured in SATIMO near field measurement system and are compared during the following setups: (a) antennas alone in the near field system, without the presence of a plastic part; (b) antennas mounted on the inside curved surface of a driver’s side mirror cover; (c) antennas mounted on the outside curved surface of the driver’s side mirror cover; (d) antennas mounted on a flat trunk lid; (e) antennas mounted on a curved plastic retrieved from the A-pillar of a vehicle. Comparison among the antennas radiation pattern measurements during these different setups, results in the conclusion that the inside surface of the side mirror cover is the most suitable position to mount the presented dual-band antennas. The curvature of the inside surface at the point where the antenna was mounted is less steep than the placement point at the outside surface, allowing the antenna to keep its polarization axis mostly unaffected. Moreover, the curve of the inside surface makes the antenna radiation more directional, creating an increase in the antenna gain. The side mirror cover, compared to trunk lid, is further from the ground protecting the antenna radiation from additional reflections.
Pavement distress classification is a vital step for automatic pavement inspection and maintenance. Recently, patch-based approaches have achieved promising performances and thus extensive attention in this field. However, these methods simply assume that all patches contribute equally to the distress classification, leading to weakly discriminating abilities of models. Moreover, their tedious processes also leads to a low efficiency in inference. In this letter, we present a novel patch-based pavement distress classification approach named Deep Patch Soft Selective Learning (DPS$^2$L), which addresses these issues. Similar to other patch-based approaches, DPS$^2$L partitions the pavement images into patches and aggregates the patch features to accomplish the task. To address the first issue, we introduce a succinct Soft Patch Feature Selection Network (SPFSN) to assess the importance of each patch to the distress classification with a score based on its feature. These scores will be considered as patch-wise weights for feature aggregation. In such a manner, the most discriminative patches are selected in a soft way, and thereby benefit the final classification. To address the inference efficiency issue, knowledge distillation is leveraged to transfer the classification knowledge from DPS$^2$L to the image-based approaches, such as EfficientNet-B3. This distilled model enables incorporating both the advantages of patch-based approaches in classification performance and the advantages of image-based approaches in inference efficiency. Extensive experiments on a large-scale pavement image dataset named CQU-BPDD demonstrates the superiority of our methods over baselines regardless of performance or efficiency.
Presented is an improved bandgap reference, which has the performance of high accuracy and can generate the required voltage reference. In this bandgap reference, the improved base current compensation is proposed to eliminate the effect of the base current. Meanwhile, a high reference voltage generator is used to provide configurable output voltages of 1.2/1.8/2.5/3.3V needed by DC-DC converters. The bandgap reference is realized in a standard 180nm CMOS process with an area of 0.05 mm×mm. Among the 5 sample chips of the reference, in the temperature range of -40 oC to 125 oC, the temperature coefficients of all the reference voltages range from 3 ppm/oC to 38 ppm/oC. The best average value of temperature coefficients is 6.03 ppm/oC when the reference voltage is 2.5V. The best line sensitivities (LS) is 0.23%/V when the reference voltage is 1.8V with the power consumption of 150μW@VDD=5V;
It is a challenging issue how to improve the accuracy of image matching in computer vision. To address this issue, an image matching method is proposed, which is via progressive priors of a putative dataset. Distance ratio priors of a putative dataset are initially employed to calculate a tentative deformation through geometric constraints. Progressive priors of the putative dataset, obtained by the tentative deformation, are then engaged to improve the accuracy of image matching by estimating a global deformation. The comparison experiments illustrate that our proposed method more effectively enhances the accuracy of image matching than six state-of-the-art methods.