To improve the accuracy of short term traffic flow prediction and to solve the problems of nonlinearity of short term traffic flow, more noise in the data, and more difficult to determine the parametes of long short term memory networks, a combined traffic flow prediction model based on variational modal decomposition (VMD) and improved dung beetle optimization-long short term memory network (IDBO-LSTM) is proposed. First, to extract various modal components, the historical traffic flow data are smoothed using variational modal decomposition (VMD). Second, the LSTM prediction model is built for each individual subsequence, and the parameters of the LSTM are optimized using the IDBO algorithm which combines Singer chaos mapping, variable spiral search strategy, and Levy flight strategy. Finally, to acquire the final prediction results, the predicted values of various subsequences are added up and reassembled. Experiments were conducted using data collected from eight sensors along an interstate highway in California, and taking the straight road morning peak (S-M) data as an example, compared with LSTM and VMD-LSTM, the MAE of VMD-IDBO-LSTM is reduced by 26.69 and 7.5108, MAPE is reduced by 8.08059% and 2.27569%, and RMSE is reduced by 33.6912 and 8.7657. According to the findings, the VMD-IDBO-LSTM model that was proposed is capable of significantly improving the accuracy of short-term traffic flow prediction while also effectively addressing nonlinearity, data noise, and the difficulty of identifying the LSTM parameters.

Satellite remote sensing technology significantly contributes to intelligent transportation by optimizing traffic planning via global perspectives and rich data, enhancing traffic efficiency and reducing environmental impact. However, current target detection models frequently exhibit low accuracy in vehicle detection tasks due to complex background interference in satellite imageries and a need for critical semantic information. To improve vehicle target detection accuracy, this study introduces SatDetX-YOLO, a vehicle detection model for satellite remote sensing images based on YOLOv8. The model involves reconstructing the backbone network with FasterNet for enhanced feature extraction, a redesigned decoupled head for improved computational efficiency and complex data processing, and incorporating the Deformable Attention Module (DAM) to increase sensitivity to small targets and feature correlation capture. Employing the Maximum Probabilistic Distance IoU (MPDIoU) loss function enhances adaptability and generalization to diverse vehicle targets. Experimental results demonstrate that under comparable FPS, SatDetX-YOLO’s Precision (P), Recall (R), and Mean Average Precision (mAP) improved by 3.5%, 3.3%, and 3.2%, respectively. Despite a minor reduction in FPS, the model significantly enhances detection accuracy, striking a balance between accuracy and speed.

The lack of vehicle feature information and the limited number of pixels in high-definition remote-sensing images causes difficulties in vehicle detection. This paper proposes U-YOLO, a vehicle detection method that integrates multi-scale features, attention mechanisms, and sub-pixel convolution. The adaptive fusion module (AF) is added to the backbone of the YOLO detection model to increase the underlying structural information of the feature map. Cross-scale channel attention (CSCA) is introduced to the feature fusion part to obtain the vehicle’s explicit semantic information and further refine the feature map. The sub-pixel convolution module (SC) is used to replace the linear interpolation up-sampling of the original model, and the vehicle target feature map is enlarged to further improve the vehicle detection accuracy. The detection accuracies on the open-source datasets NWPU VHR-10 and DOTA were 91.35% and 71.38%. Compared with the original network model, the detection accuracy on these two datasets was increased by 6.89% and 4.94%, respectively. Compared with the classic target detection networks commonly used in RFBnet, M2det, and SSD300, the average accuracy rate values increased by 6.84%, 6.38%, and 12.41%, respectively. The proposed method effectively solves the problem of low vehicle detection accuracy. It provides an effective basis for promoting the application of high-definition remote-sensing images in traffic target detection and traffic flow parameter detection.

In this paper, we address the issues of insufficient accuracy and frequent identity switching in the multi-target tracking algorithm DeepSORT by proposing two improvement strategies. First, we optimize the appearance feature extraction process by training a lightweight appearance extraction network (OSNet) on a vehicle re-identification dataset. This makes the appearance features better suited for the vehicle tracking model required in our paper. Second, we improve the metric of motion features by using the original IOU distance metric or GIOU metrics. The optimized tracking algorithm using GIOU achieves effective improvements in tracking precision and accuracy. The experimental results show that the improved vehicle tracking models MOTA and IDF1 are enhanced by 4.6% and 5.9%, respectively. This allows for the stable tracking of vehicles and reduces the occurrence of identity switching phenomenon to a certain extent.

Raw material inventory control is indispensable for ensuring the cost reduction and efficiency of enterprises. Silica powder is an essential raw material for new energy enterprises. The inventory control of silicon powder is of great concern to enterprises, but due to the complexity of the market environment and the inadequacy of information technology, inventory control of silica powder has been ineffective. One of the most significant reasons for this is that existing methods encounter difficulty in effectively extracting the local and long-term characteristics of the data, which leads to significant errors in forecasting and poor accuracy. This study focuses on improving the accuracy of corporate inventory forecasting. We propose an improved CNN-BiLSTM-attention prediction model that uses convolutional neural networks (CNNs) to extract the local features from a dataset. The attention mechanism (attention) uses the point multiplication method to weigh the acquired features and the bidirectional long short-term memory (BiLSTM) network to acquire the long-term features of the dataset. The final output of the model is the predicted value of silica powder and the evaluation metrics. The proposed model is compared with five other models: CNN, LSTM, CNN-LSTM, CNN-BiLSTM, and CNN-LSTM-attention. The experiments show that the improved CNN-BiLSTM-attention prediction model can predict inbound and outbound silica powder very well. The accuracy of the prediction of the inbound test set is higher than that of the other five models by 7.429%, 11.813%, 15.365%, 10.331%, and 5.821%, respectively. The accuracy of the outbound storage prediction is higher than that of the other five models by 14.535%, 15.135%, 1.603%, 7.584%, and 18.784%, respectively.

The randomness of passengers’ travel and the blindness of empty drivers seeking passengers can lead to a serious imbalance in the spatio-temporal distribution of taxi supply and demand. In order to realize the accurate prediction of taxi demand, promote a balance between taxi supply and demand, and respond to the requirements of the sustainable development of urban transportation, a travel demand prediction model based on Sparrow Search Algorithm incorporating sine-cosine and Cauchy variants (SCSSA), Convolutional Neural Network (CNN), and Bi-directional Long Short-Term Memory (BiLSTM) is proposed. The key factors affecting travel demand are identified by constructing a set of influencing factors for feature correlation analysis. In order to overcome the overfitting or underfitting phenomenon caused by the improper parameter configuration of the CNN-BiLSTM model, the SCSSA algorithm is utilized to optimize the model. By fine tuning the model parameters, the algorithm enhanced the model’s adaptability to dataset characteristics and improved the accuracy of the prediction results. Compared with CNN, LSTM, CNN- LSTM, CNN-BiLSTM, and SSA-CNN-BiLSTM models, the Root Mean Square Error is decreased by 10.77 on average.