An Environmentally Sustainable Software-Defined Networking Data Dissemination Method for Mixed Traffic Flows in RSU Clouds with Energy Restriction.

The connected multi road side unit (RSU) environment can be envisioned as the RSU cloud. In this paper, the Software-Defined Networking (SDN) framework is utilized to dynamically reconfigure the RSU clouds for the mixed traffic flows with energy restrictions, which are composed of five categories of vehicles with distinctive communication demands. An environmentally sustainable SDN data dissemination method for safer and greener transportation solutions is thus proposed, aiming to achieve the lowest overall SDN cloud delay with the least working hosts and minimum energy consumption, which is a mixed integer linear programming problem (MILP). To solve the problem, Joint optimization algorithms with Finite resources (JF) in three hyperparameters versions, JF (DW = 0.3, HW = 0.7), JF (DW = 0.5, HW = 0.5) and JF (DW = 0.7, HW = 0.3), were proposed, which are in contrast with single-objective optimization algorithms, the Host Optimization (H) algorithm, and the Delay optimization (D) algorithm. Results show that JF (DW = 0.3, HW = 0.7) and JF (DW = 0.5, HW = 0.5), when compared with the D algorithm, usually had slightly larger cloud delays, but fewer working hosts and energy consumptions, which has vital significance for enhancing energy efficiency and environmental protection, and shows the superiority of JFs over the D algorithm. Meanwhile, the H algorithm had the least working hosts and fewest energy consumptions under the same conditions, but completely ignored the explosive surge of delay, which is not desirable for most cases of the SDN RSU cloud. Further analysis showed that the larger the network topology of the SDN cloud, the harder it was to find a feasible network configuration. Therefore, when designing an environmentally sustainable SDN RSU cloud for the greener future mobility of intelligent transportation systems, its size should be limited or partitioned into a relatively small topology.

Infrared Sensation-Based Salient Targets Enhancement Methods in Low-Visibility Scenes.

Thermal imaging is an important technology in low-visibility environments, and due to the blurred edges and low contrast of infrared images, enhancement processing is of vital importance. However, to some extent, the existing enhancement algorithms based on pixel-level information ignore the salient feature of targets, the temperature which effectively separates the targets by their color. Therefore, based on the temperature and pixel features of infrared images, first, a threshold denoising model based on wavelet transformation with bilateral filtering (WTBF) was proposed. Second, our group proposed a salient components enhancement method based on a multi-scale retinex algorithm combined with frequency-tuned salient region extraction (MSRFT). Third, the image contrast and noise distribution were improved by using salient features of orientation, color, and illuminance of night or snow targets. Finally, the accuracy of the bounding box of enhanced images was tested by the pre-trained and improved object detector. The results show that the improved method can reach an accuracy of 90% of snow targets, and the average precision of car and people categories improved in four low-visibility scenes, which demonstrates the high accuracy and adaptability of the proposed methods of great significance for target detection, trajectory tracking, and danger warning of automobile driving.

An Online Classification Method for Fault Diagnosis of Railway Turnouts.

Railway turnout system is a key infrastructure to railway safety and efficiency. However, it is prone to failure in the field. Therefore, many railway departments have adopted a monitoring system to monitor the operation status of turnouts. With monitoring data collected, many researchers have proposed different fault-diagnosis methods. However, many of the existing methods cannot realize real-time updating or deal with new fault types. This paper-based on imbalanced data-proposes a Bayes-based online turnout fault-diagnosis method, which realizes incremental learning and scalable fault recognition. First, the basic conceptions of the turnout system are introduced. Next, the feature extraction and processing of the imbalanced monitoring data are introduced. Then, an online diagnosis method based on Bayesian incremental learning and scalable fault recognition is proposed, followed by the experiment with filed data from Guangzhou Railway. The results show that the scalable fault-recognition method can reach an accuracy of 99.11%, and the training time of the Bayesian incremental learning model reduces 29.97% without decreasing the accuracy, which demonstrates the high accuracy, adaptability and efficiency of the proposed model, of great significance for labor-saving, timely maintenance and further, safety and efficiency of railway transportation.

Time-of-day Control Double-Order Optimization of Traffic Safety and Data-Driven Intersections.

This paper proposes a novel two-order optimization model of the division of time-of-day control segmented points of road intersection to address the limitations of the randomness of artificial experience, avoid the complex multi-factor division calculation, and optimize the traditional model over traffic safety and data-driven methods. For the first-order optimization-that is, deep optimization of the model input data-we first increase the dimension of traditional traffic flow data by data-driven and traffic safety methods, and develop a vector quantity to represent the size, direction, and time frequency with conflict point traffic of the total traffic flow at a certain intersection for a period by introducing a 3D vector of intersection traffic flow. Then, a time-series segmentation algorithm is used to recurse the distance amongst adjacent vectors to obtain the initial scheme of segmented points, and the segmentation points are finally divided by the combination of the preliminary scheme. For the second-order optimization-that is, model adaptability analysis-the traffic flow data at intersections are subjected to standardised processing by five-number summary. The different traffic flow characteristics of the intersection are categorised by the K central point clustering algorithm of big data, and an applicability analysis of each type of intersection is conducted by using an innovated piecewise point division model. The actual traffic flow data of 155 intersections in Yuecheng District, Shaoxing, China, in 2016 are tested. Four types of intersections in the tested range are evaluated separately by the innovated piecewise point division model and the traditional total flow segmentation model on the basis of Synchro 7 simulation software. It is shown that when the innovated double-order optimization model is used in the intersection according to the 'hump-type' traffic flow characteristic, its control is more accurate and efficient than that of the traditional total flow segmentation model. The total delay time is reduced by approximately 5.6%. In particular, the delay time in the near-peak-flow buffer period is significantly reduced by approximately 17%. At the same time, the traffic accident rate has also dropped significantly, effectively improving traffic safety at intersections.