Can we trust our eyes? Interpreting the misperception of road safety from street view images and deep learning.

Road safety is a critical concern that impacts both human lives and urban development, drawing significant attention from city managers and researchers. The perception of road safety has gained increasing research interest due to its close connection with the behavior of road users. However, safety isn't always as it appears, and there is a scarcity of studies examining the association and mismatch between road traffic safety and road safety perceptions at the city scale, primarily due to the time-consuming nature of data acquisition. In this study, we applied an advanced deep learning model and street view images to predict and map human perception scores of road safety in Manhattan. We then explored the association and mismatch between these perception scores and traffic crash rates, while also interpreting the influence of the built environment on this disparity. The results showed that there was heterogeneity in the distribution of road safety perception scores. Furthermore, the study found a positive correlation between perception scores and crash rates, indicating that higher perception scores were associated with higher crash rates. In this study, we also concluded four perception patterns: "Safer than it looks", "Safe as it looks", "More dangerous than it looks", and "Dangerous as it looks". Wall view index, tree view index, building view index, distance to the nearest traffic signals, and street width were found to significantly influence these perception patterns. Notably, our findings underscored the crucial role of traffic lights in the "More dangerous than it looks" pattern. While traffic lights may enhance people's perception of safety, areas in close proximity to traffic lights were identified as potentially accident-prone regions.

A comprehensive study of macro factors related to traffic fatality rates by XGBoost-based model and GIS techniques.

With the fast development of economics, road safety is becoming a serious problem. Exploring macro factors is effective to improve road safety. However, the existing studies have some limitations: (1) The existing studies only considered one aspect of macro factors and constructed models based on a few data samples. (2) The methods commonly used cannot address the non-linear relationship or calculate the feature importance. The findings obtained from such models may be limited and biased. To address the limitations, this study proposes a BO-CV-XGBoost framework to explore the macro factors related to traffic fatality rate classes based on a high-dimensional dataset that fully considers the impact of multi-factor interaction with adequate data samples. The proposed framework is applied to a dataset in the US. 453 county-level macro factors are collected from various data sources, covering ten macro aspects, including topography, transportation, etc. The optimized BO-CV-XGBoost model obtains the best classification performance with an AUC of 0.8977 and an accuracy of 85.02%. Compared with other methods, the proposed model has superiority on fatality rate classification. Ten macro factors are identified, including 'Current-dollar GDP', 'highway miles per person', etc. The ten factors contain four aspects of information, including economics, transportation, education, and medical condition. Geographic information system (GIS) techniques are further used for spatial analysis of the identified macro factors. Therefore, targeted and effective measures are accordingly proposed to prevent traffic fatalities and improve road safety.

Analysis of motorcycle accidents using association rule mining-based framework with parameter optimization and GIS technology.

INTRODUCTION: Analyzing key factors of motorcycle accidents is an effective method to reduce fatalities and improve road safety. Association Rule Mining (ARM) is an efficient data mining method to identify critical factors associated with injury severity. However, the existing studies have some limitations in applying ARM: (a) Most studies determined parameter thresholds of ARM subjectively, which lacks objectiveness and efficiency; (b) Most studies only listed rules with high parameter thresholds, while lacking in-depth analysis of multiple-item rules. Besides, the existing studies seldom conducted a spatial analysis of motorcycle accidents, which can provide intuitive suggestions for policymakers. METHOD: To address these limitations, this study proposes an ARM-based framework to identify critical factors related to motorcycle injury severity. A method for parameter optimization is proposed to objectively determine parameter thresholds in ARM. A method of factor extraction is proposed to identify individual key factors from 2-item rules and boosting factors from multiple-item rules. Geographic information system (GIS) is adopted to explore the spatial relationship between key factors and motorcycle injury severity. RESULTS AND CONCLUSIONS: The framework is applied to a case study of motorcycle accidents in Victoria, Australia. Fifteen attributes are selected after data preprocessing. 0.03 and 0.7 are determined as the best thresholds of support and confidence in ARM. Five individual key factors and four boosting factors are identified to be related to fatal injury. Spatial analysis is conducted by GIS to present hot spots of motorcycle accidents. The proposed framework has been validated to have better performance on parameter optimization and rule analysis in ARM. Practical applications: The hot spots of motorcycle accidents related to fatal factors are presented in GIS maps. Policymakers can refer to those maps straightforwardly when decision making. This framework can be applied to various kinds of traffic accidents to improve the performance of severity analysis.

A long short-term memory-based framework for crash detection on freeways with traffic data of different temporal resolutions.

Traffic crash detection is a major component of intelligent transportation systems. It can explore inner relationships between traffic conditions and crash risk, prevent potential crashes, and improve road safety. However, there exist some limitations in current studies on crash detection: (1) The commonly used machine learning methods cannot simulate the evolving transitions of traffic conditions before crash occurrences; (2) Current models collected traffic data of only one temporal resolution, which cannot fully represent traffic trends in different time intervals. Therefore, this study proposes a Long short-term memory (LSTM) based framework considering traffic data of different temporal resolutions (LSTMDTR) for crash detection. LSTM is an effective deep learning method to capture the long-term dependency and dynamic transitions of pre-crash conditions. Three LSTM networks considering traffic data of different temporal resolutions are constructed, which can comprehensively indicate traffic variations in different time intervals. A fully-connected layer is used to combine the outputs of three LSTM networks, and a dropout layer is used to reduce overfitting and improve prediction performance. The LSTMDTR model is implemented on datasets of I880-N and I805-N in California, America. The results indicate that the LSTMDTR model can obtain satisfactory performance on crash detection, with the highest crash accuracy of 70.43 %. LSTMDTR models constructed on one freeway can be transferred to other similar freeways, with 65.12 % of crash accuracy on transferability. Compared with machine learning methods and LSTM models with one or two temporal resolutions, the LSTMDTR model has been validated to perform better on crash detection and transferability. A proper number of neurons in the LSTMDTR model should be determined in real applications considering acceptable detection performance and computation time. The dropout technique can reduce overfitting and improve the generalization ability of the LSTMDTR model, increasing crash accuracy from 64.49 % to 70.43 %.

Soft detection of 5-day BOD with sparse matrix in city harbor water using deep learning techniques.

To better control and manage harbor water quality is an important mission for coastal cities such as New York City (NYC). To achieve this, managers and governors need keep track of key quality indicators, such as temperature, pH, and dissolved oxygen. Among these, the Biochemical Oxygen Demand (BOD) over five days is a critical indicator that requires much time and effort to detect, causing great inconvenience in both academia and industry. Existing experimental and statistical methods cannot effectively solve the detection time problem or provide limited accuracy. Also, due to various human-made mistakes or facility issues, the data used for BOD detection and prediction contain many missing values, resulting in a sparse matrix. Few studies have addressed the sparse matrix problem while developing statistical detection methods. To address these gaps, we propose a deep learning based model that combines Deep Matrix Factorization (DMF) and Deep Neural Network (DNN). The model was able to solve the sparse matrix problem more intelligently and predict the BOD value more accurately. To test its effectiveness, we conducted a case study on the NYC harbor water, based on 32,323 water samples. The results showed that the proposed method achieved 11.54%-17.23% lower RMSE than conventional matrix completion methods, and 19.20%-25.16% lower RMSE than traditional machine learning algorithms.

Methoxo[N'-(3-meth-oxy-2-oxidobenzyl-idene)benzohydrazidato]oxidovanadium(V).

In the title complex, [V(C(15)H(12)N(2)O(4))(CH(3)O)O], the V(V) ion exhibits a distorted square-pyramidal coordination geometry; three donor atoms from a hydrazone ligand and one O atom of the deprotonated methanol define the coordination basal plane. The V(V) ion is displaced by 0.464 (1) Šfrom the basal plane towards the axial oxide O atom. Intra-molecular O-H⋯N hydrogen bonding occurs. Inter-molecular C-H⋯O hydrogen bonding is also observed in the crystal structure.