Accurate traffic speed forecasting not only can help traffic management departments make better judgments and improve the efficacy of road monitoring but also can help drivers plan their driving routes and arrive safely and smoothly at their destination. This paper focuses on the lack of traffic speed data and proposes a method for traffic speed forecasting based on the multitemporal traffic flow volume of the previous and later moment states. First, according to traffic flow volume data, the different traffic patterns of previous and later moment states were extracted. Second, the performance of five forecasting models, namely, long short-term memory (LSTM), backpropagation (BP), classification and regression trees, k-nearest neighbor, and support vector regression, were compared. Finally, the model with the best prediction results was used to conduct sensitivity analysis experiments for different traffic patterns. Through a real-data case study, we found that the LSTM model has the highest prediction accuracy compared to other models in both time and space. This traffic pattern "previous = 3 and later = 3" can forecast traffic speed more accurately, and its forecasting ability is robust across a range of scenarios.

Traffic signs are one of the main carriers of road information, the reflectivity of traffic signs at night directly affects the driver's access to road information. Fast and accurate traffic sign reflectivity detection method is conducive to the efficient operation of road maintenance operations to protect the lives of motorists. This paper proposed a novel method for detecting the reflectivity of traffic signs, which proposed the concept of sign’s luminance value. By collecting photos of traffic signs on the road at night through the detection vehicle, using Yolov4 algorithm and Deeplabv3 algorithm for sign target detection and image segmentation respectively, then the signs were converted into grayscale images after grayscale processing, and the difference between the grayscale value of the sign and its background was calculated to obtain the luminance value of the sign, and the luminance value was used to represent the reflective performance of the sign. This method was used to conduct the actual test experiment of road traffic sign reflectivity, and compared with the sign reflectivity results obtained by the traditional retroreflective coefficient detection method. Through data analysis, it was found that the test results of this method have significant correlation with the test results of the traditional method. The traffic sign reflectivity detection method proposed in this paper obtains more accurate detection results and avoids the disadvantages of the traditional retroreflective coefficient detection method, such as complicated operation, high instrument requirements, low detection efficiency, and failure to consider the impact of the actual installation position of the sign on the driver's visual recognition, and is more suitable for engineering practice.

No abstract

No abstract