Search

Steel cord conveyor belt has the advantages of high tensile strength, high friction resistance and strong impact resistance. It is widely used in coal, mining, port, electric power and other fields...

The emotion recognition with electroencephalography (EEG) has been widely studied using the deep learning methods, but the topology of EEG channels is rarely exploited completely. In this paper, we propose a self-attention coherence clustering based on multi-pooling graph convolutional network (SCC-MPGCN) model for EEG emotion recognition. The adjacency matrix is constructed based on phase-locking value to describe the intrinsic relationship between different EEG electrodes as graph signals. The graph Laplacian matrix is obtained from the adjacency matrix and then is fed into the graph convolutional layers to learn the generalized features. Moreover, we propose a novel graph coarsening method called SCC, using the coherence to cluster the nodes. The benefits are that the dimensionality of adjacency matrix can be reduced and the global information can be achieved from the raw data. Meanwhile, a MPGCN block is introduced to learn the generalized features of emotional states. The fully-connected layer and a softmax layer are adopted to derive the final classification results. We carry out the extensive experiments on DEAP dataset and the results show that the proposed method has better classification results than the state-of-the-art methods with the ten-fold cross-validation. And the model achieves the emotion recognition performance with a mean accuracy of 96.37%, 97.02%, 96.72% on valence, arousal, and dominance dimension, respectively.

Conveyor belt is an important element of the conveyor. The strength of the steel cord conveyor belt largely determines the carrying capacity, and it also has a great impact on operational safety. In this paper, the effect of different factors on the strength of the steel cord conveyor belt splices was studied. The FEM was used for simulation analysis, and the corresponding tensile experiments were carried out to verify. Steel cords of different lengths were simulated, and the simulation results agree well with the experimental results. On this basis, the steel cord length, steel cord diameter, rubber thickness, and different number of steel cords were investigated to study the effect on the pullout force of the steel cord conveyor belt splice. The numerical simulation results show that different steel cord diameters have more significant effect on the strength of the conveyor belt splice compared to rubber. The steel cord length and steel cord diameter impact on the steel cord conveyor belt is approximately linear. For the different number of the steel cords, the increase in the number of steel cords does not mean that the tension will increase by the same multiple, and the increase in pullout force is less than the increase in the number of steel cords. It provides guidance for the production of steel cord conveyor belts.

LNG storage tank is an important storage and measurement tool for LNG, but there is no relevant standard for LNG storage tank capacity measurement.In this paper, a method of measuring the capacity of LNG cryogenic storage tank based on laser measurement technology is proposed. According to the geometry of LNG tank and the characteristics of low temperature operation, a mathematical model is established to solve the problem of LNG tank capacity measurement, which has good practical value.

In order to solve the problem of dizziness caused by the inconsistency between visual and somatosensory movements of people in a virtual reality system, a set of stride detection device composed of $\mathbf{2}\times \mathbf{11}$ array of foot pressure sensors was designed. The piezoresistive characteristics and rebound time of the foot pressure sensor were studied. A footstep pressure signal detection circuit is designed, then the footstep pressure signal through the MCU is collected, and calculated the pace. At the same time, the ARIMA model is used to predict speed changes. The experimental results show that the piezoresistive characteristics of the foot pressure sensor meet the requirements of foot detection, and the rebound time is less than 170 ms; the pace detection device detects the step detection accuracy and prediction accuracy are both greater than 95%, which meets the design requirements of VR motion platforms.

Vehicle detection technology is the key technology of intelligent transportation systems, attracting the attention of many researchers. Although much literature has been published concerning daytime vehicle detection, little has been published concerning nighttime vehicle detection. In this study, a nighttime vehicle detection algorithm, consisting of headlight segmentation, headlight pairing and headlight tracking, is proposed. First, the pixels of the headlights are segmented in nighttime traffic images, through the use of the thresholding method. Then the pixels of the headlights are grouped and labeled, to analyze the characteristics of related components, such as area, location and size. Headlights are paired based on their location and size and then tracked via a tracking procedure designed to detect vehicles. Vehicles with only one headlight or those with three or four headlights are also detected. Experimental results show that the proposed algorithm is robust and effective in detecting vehicles in nighttime traffic.

The off-chip memory bandwidth is often a critical part of a video processing system. Traditional cache usually behaves poor since some video algorithms, such as motion compensation, tend to access memory in an inefficient way. An efficient memory bandwidth architecture for video decoder is proposed in this paper. Two bandwidth optimization strategies are proposed through pixels reconstructed and data arrangements without any degradation of picture quality. The experimental results show that two strategies succeed in reducing external memory bandwidth of motion compensation by 30% and the data arrangements strategy can solve the latency problem better than the previous solutions. The proposed algorithm is also computationally simple, reducing latency time and creating sufficient accessibility.

Audio artifacts are inconsistencies in the time-frequency domain caused by the compression rules of an encoding algorithm. It is well known, audio artifacts can affect the quality of audio sharply. This paper proposes a novel postprocessing technique for reducing audio artifacts. Compared with the original methods, the proposed method reduces audio error data not only in time domain but also in frequency domain. Based on the experimental results, the proposed post-filtering technique provides satisfactory performance and consistently works well for a wide variety of audio.