A study on expression recognition based on improved mobilenetV2 network.

This paper proposes an improved strategy for the MobileNetV2 neural network(I-MobileNetV2) in response to problems such as large parameter quantities in existing deep convolutional neural networks and the shortcomings of the lightweight neural network MobileNetV2 such as easy loss of feature information, poor real-time performance, and low accuracy rate in facial emotion recognition tasks. The network inherits the characteristics of MobilenetV2 depthwise separated convolution, signifying a reduction in computational load while maintaining a lightweight profile. It utilizes a reverse fusion mechanism to retain negative features, which makes the information less likely to be lost. The SELU activation function is used to replace the RELU6 activation function to avoid gradient vanishing. Meanwhile, to improve the feature recognition capability, the channel attention mechanism (Squeeze-and-Excitation Networks (SE-Net)) is integrated into the MobilenetV2 network. Experiments conducted on the facial expression datasets FER2013 and CK + showed that the proposed network model achieved facial expression recognition accuracies of 68.62% and 95.96%, improving upon the MobileNetV2 model by 0.72% and 6.14% respectively, and the parameter count decreased by 83.8%. These results empirically verify the effectiveness of the improvements made to the network model.

Preparation of aligned nanofibers using parallel inductive-plates assisted electrospinning.

Electrospinning is a simple, cost-effective, and versatile technique for fabrication of nanofibers. However, nanofibers obtained from the conventional electrospinning are typically disordered, which seriously limits their application. In this work, we present a novel and facile technique to obtain aligned nanofibers with high efficiency by using parallel inductive-plates assisted electrospinning (PIES). In this new electrospinning setup, the electrostatic spinneret is contained in a pair of parallel inductive-plates, which can change the shape and direction of the electric field line during the electrospinning so as to control the flight trajectory and spatial alignment of the spinning nanofibers. This electrospinning setup can divide the electric field line into two parts which are respectively directed to the edge of the upper and lower inductive-plates. Then the nanofibers move along the electric field line, suspend and align between the parallel inductive-plates. Finally, the well aligned nanofibers could be easily transferred onto other substrates for further characterizations and applications. The aligned nanofibers with an average diameter of 469 ± 115 nm and a length as long as 140 mm were successfully achieved by using PIES technique. Moreover, nanofiber arrays with different cross angles and three-dimensional films formed by the aligned nanofibers were also facilely obtained. The novel PIES developed in this work has been proved to be a facile, cost-effective and promising approach to prepare aligned nanofibers for a wide range of applications.