Search

ABSTRACTLow visibility in foggy weather can easily cause accidents and affect normal life. Accurate visibility estimate is critical to transportation, aviation, marine and other fields. For visibility estimation, this paper proposes a visibility estimation algorithm Mean Square Error Based on Convolution Neural Network (MSEBCNN) based on continuous surveillance video, which is an improved neural network deep learning algorithm. In this paper, the images of airport surveillance video were extracted every 15 s and the Region of Interest (ROI) in each image was selected as the measurement area. The improved Convolution Neural Network (CNN) was constructed by replacing the Softmax layer of the last layer of the convolutional neural network with the Mean Square Error (MSE) layer of the objective regression function, and the gradient descent method was used for fitting training to achieve the continuous prediction of visibility. By experiment, the accuracy of prediction at a range of 0–500 m is 97.51%, while the accuracy of prediction at a range of 500–1600 m is 91.67%, which obtained a high prediction accuracy.

The use of dead reckoning and fingerprint matching for navigation is a widespread technical method. However, fingerprint mismatching and low fusion accuracy are prevalent issues in indoor navigation systems. This work presents an improved dynamic time warping and a chicken particle filter to handle these two challenges. To generate the Horizontal and Vertical (HV) fingerprint, the pitch and roll are employed instead of the original fingerprint intensity to extract the horizontal and vertical components of the magnetic field fingerprint. Derivative dynamic time warping employs the HV fingerprint in its derivative form, which receives higher-level features because of the consideration of fingerprint shape information. Chicken Swarm Optimization (CSO) is used to enhance particle weights, which minimizes position error to tackle the particle impoverishment problem for a fusion navigation system. The results of the experiments suggest that the enhanced algorithm can improve indoor navigation accuracy significantly.

•To solve the problem that BP neural network is easy to fall into local optimum.•BP neural network based on levy particle swarm optimization is proposed to estimate battery state of charge.•The accuracy of the algorithm is verified by using industry public data sets.•It is valuable for electric vehicles.