Artificial intelligence,such as deep learning technology,has advanced the study of facial expression recognition since facial expression carries rich emotional information and is significant for many naturalistic situ...Artificial intelligence,such as deep learning technology,has advanced the study of facial expression recognition since facial expression carries rich emotional information and is significant for many naturalistic situations.To pursue a high facial expression recognition accuracy,the network model of deep learning is generally designed to be very deep while the model’s real-time performance is typically constrained and limited.With MobileNetV3,a lightweight model with a good accuracy,a further study is conducted by adding a basic ResNet module to each of its existing modules and an SSH(Single Stage Headless Face Detector)context module to expand the model’s perceptual field.In this article,the enhanced model named Res-MobileNetV3,could alleviate the subpar of real-time performance and compress the size of large network models,which can process information at a rate of up to 33 frames per second.Although the improved model has been verified to be slightly inferior to the current state-of-the-art method in aspect of accuracy rate on the publically available face expression datasets,it can bring a good balance on accuracy,real-time performance,model size and model complexity in practical applications.展开更多
The current method of estimating the fatigue life of railway structures is to calculating the equivalent stress amplitude based on the measured stress data. However, the random of the measured data is not considered. ...The current method of estimating the fatigue life of railway structures is to calculating the equivalent stress amplitude based on the measured stress data. However, the random of the measured data is not considered. In this paper, a new method was established to compute the equivalent stress amplitude to evaluate the fatigue damage based on the measurable randomness, since the equivalent stress is the key parameter for assessment of structure fatigue life and load derivation. The equivalent stress amplitude of a high-speed train welded bogie frame was found to obey normal distribution under uniform operation route that verified by on-track dynamic stress data, and the proposed model is, in effect, an improved version of the mathematical model used to calculate the equivalent stress amplitude. The data of a long-term, on-track dynamic stress test program was analyzed to find that the normal distribution parameters of equivalent stress amplitude values differ across different operation route. Thus, the fatigue damage of the high-speed train welded bogie frame can be evaluated by the proposed method if the running schedule of the train is known a priori. The results also showed that the equivalent stress amplitude of the region connected to the power system is more random than in other regions of the bogie frame.展开更多
In this paper,a fast-speed and real-time online rail inspection method based on half-cycle orthogonal power demodulation algorithm is proposed.For this method,the power characters of de-tection signal which represent ...In this paper,a fast-speed and real-time online rail inspection method based on half-cycle orthogonal power demodulation algorithm is proposed.For this method,the power characters of de-tection signal which represent the degree of rail track can be calculated using only half-cycle detec-tion signal because of the symmetry characteristic of detected sine signal and reference signal.The theoretical analysis,simulation results and experiment results show that the demodulation preci-sion of proposed method is almost equal to fast Fourier transform(FFT)demodulation method and orthogonal demodulation method,but has high demodulation efficiency and less FPGA resources cost.A high-speed experiment system based on three coils structured sensor is built for rail inspec-tion experiment at a moving speed of 200 km/h.The experiment results show that proposed meth-od is more effective for rail inspection and the time resolution of proposed method is double of clas-sic method that based on FFT and orthogonal.展开更多
基金supported by China Academy of Railway Sciences Corporation Limited(No.2021YJ127).
文摘Artificial intelligence,such as deep learning technology,has advanced the study of facial expression recognition since facial expression carries rich emotional information and is significant for many naturalistic situations.To pursue a high facial expression recognition accuracy,the network model of deep learning is generally designed to be very deep while the model’s real-time performance is typically constrained and limited.With MobileNetV3,a lightweight model with a good accuracy,a further study is conducted by adding a basic ResNet module to each of its existing modules and an SSH(Single Stage Headless Face Detector)context module to expand the model’s perceptual field.In this article,the enhanced model named Res-MobileNetV3,could alleviate the subpar of real-time performance and compress the size of large network models,which can process information at a rate of up to 33 frames per second.Although the improved model has been verified to be slightly inferior to the current state-of-the-art method in aspect of accuracy rate on the publically available face expression datasets,it can bring a good balance on accuracy,real-time performance,model size and model complexity in practical applications.
基金Supported by National Key Research and Development Program of China(Grant No.2018YFB1201704)National Natural Science Foundation of China(Major Program,Grant No.11790281)
文摘The current method of estimating the fatigue life of railway structures is to calculating the equivalent stress amplitude based on the measured stress data. However, the random of the measured data is not considered. In this paper, a new method was established to compute the equivalent stress amplitude to evaluate the fatigue damage based on the measurable randomness, since the equivalent stress is the key parameter for assessment of structure fatigue life and load derivation. The equivalent stress amplitude of a high-speed train welded bogie frame was found to obey normal distribution under uniform operation route that verified by on-track dynamic stress data, and the proposed model is, in effect, an improved version of the mathematical model used to calculate the equivalent stress amplitude. The data of a long-term, on-track dynamic stress test program was analyzed to find that the normal distribution parameters of equivalent stress amplitude values differ across different operation route. Thus, the fatigue damage of the high-speed train welded bogie frame can be evaluated by the proposed method if the running schedule of the train is known a priori. The results also showed that the equivalent stress amplitude of the region connected to the power system is more random than in other regions of the bogie frame.
基金supported by the National Natural Science Found-ation of China(No.61771041).
文摘In this paper,a fast-speed and real-time online rail inspection method based on half-cycle orthogonal power demodulation algorithm is proposed.For this method,the power characters of de-tection signal which represent the degree of rail track can be calculated using only half-cycle detec-tion signal because of the symmetry characteristic of detected sine signal and reference signal.The theoretical analysis,simulation results and experiment results show that the demodulation preci-sion of proposed method is almost equal to fast Fourier transform(FFT)demodulation method and orthogonal demodulation method,but has high demodulation efficiency and less FPGA resources cost.A high-speed experiment system based on three coils structured sensor is built for rail inspec-tion experiment at a moving speed of 200 km/h.The experiment results show that proposed meth-od is more effective for rail inspection and the time resolution of proposed method is double of clas-sic method that based on FFT and orthogonal.
