Quantifying visual road environment to establish a speeding prediction model: An examination using naturalistic driving data

Naturalistic driving analysis of situational, behavioral, and psychosocial determinants of speeding

The present analysis used full-trip naturalistic driving data along with driver behavioral and psychosocial surveys to understand the individual and contextual predictors of speeding. The data were collected over a three-week period from 44 drivers and contain 3,798 full trips, with drivers speeding 7.8 % of the time. Speeding events were identified as periods when participants traveled at a velocity greater than five mph over the speed limit for at least five seconds. Data were analyzed using the Comprehensive Driver Profile (CDP) framework which uses principal component analysis (dimensionality reduction), random forest (predictive modeling), k-means clustering (grouping and profiling), and bootstrapping (profile stability) to decompose the predictive variables and driver characteristics. The final dataset included 188 candidate independent variables from the CDP framework and one dependent variable (speeding). Nine variables emerged as significant predictors of speeding onset with an AUC of 0.88, including the percent of trip time spent idling and speeding, highway driving in low traffic conditions, and positive attitudes toward phone use. Percent of trip speeding was associated with a higher likelihood of speeding by up to 42 percent, and percent trip idling was associated with it by up to 30 percent. Driver profile clusters revealed four types: Traffic & Idling Speeders, Infrequent Speeders, Frequent Speeders, and Situational Speeders. The present analysis demonstrates the importance of situational factors and individual differences in motivating speeding behavior. Countermeasures targeting speeding may be more effective if they address the root causes of the behavior in addition to the behavior itself.

The dynamic-static dual-branch deep neural network for urban speeding hotspot identification using street view image data

The visual information regarding the road environment can influence drivers' perception and judgment, often resulting in frequent speeding incidents. Identifying speeding hotspots in cities can prevent potential speeding incidents, thereby improving traffic safety levels. We propose the Dual-Branch Contextual Dynamic-Static Feature Fusion Network based on static panoramic images and dynamically changing sequence data, aiming to capture global features in the macro scene of the area and dynamically changing information in the micro view for a more accurate urban speeding hotspot area identification. For the static branch, we propose the Multi-scale Contextual Feature Aggregation Network for learning global spatial contextual association information. In the dynamic branch, we construct the Multi-view Dynamic Feature Fusion Network to capture the dynamically changing features of a scene from a continuous sequence of street view images. Additionally, we designed the Dynamic-Static Feature Correlation Fusion Structure to correlate and fuse dynamic and static features. The experimental results show that the model has good performance, and the overall recognition accuracy reaches 99.4%. The ablation experiments show that the recognition effect after the fusion of dynamic and static features is better than that of static and dynamic branches. The proposed model also shows better performance than other deep learning models. In addition, we combine image processing methods and different Class Activation Mapping (CAM) methods to extract speeding frequency visual features from the model perception results. The results show that more accurate speeding frequency features can be obtained by using LayerCAM and GradCAM-Plus for static global scenes and dynamic local sequences, respectively. In the static global scene, the speeding frequency features are mainly concentrated on the buildings and green layout on both sides of the road, while in the dynamic scene, the speeding frequency features shift with the scene changes and are mainly concentrated on the dynamically changing transition areas of greenery, roads, and surrounding buildings. The code and model used for identifying hotspots of urban traffic accidents in this study are available for access: https://github.com/gwt-ZJU/DCDSFF-Net.

Exploring road safety using alignment perspective features in real driving images: A case study on mountain freeways

INTRODUCTION

While driving, drivers frequently adapt their driving behaviors according to their perception of the road's alignment features. However, traditional two-dimensional alignment methods lack the ability to capture these features from the driver's perspective.

METHOD

This study introduces a novel method for road alignment recognition, employing image recognition technology to extract alignment perspective features, namely alignment perspective skewness (APS) and alignment perspective kurtosis (APK), from in-real driving images. Subsequently, the K-means clustering algorithm is utilized for road segment classification based on APS and APK indicators. Various sliding step length for clustering are employed, with step length ranging from 100m to 400m. Furthermore, the accident rates for different segment clusters are analyzed to explore the relationship between alignment perspective features and traffic safety. A 150 km mountain road section of the Erlianhaote-Guangzhou freewway from Huaiji to Sihui is selected as a case study.

RESULTS

The results demonstrate that using alignment perspective features as classification criteria produces favorable clustering outcomes, with superior clustering performance achieved using shorter segment lengths and fewer cluster centers. The road segment classification based on alignment perspective features reveals notable differences in accident rates across categories; while traditional two-dimensional parameters-based classification methods fail to capture these differences. The most significant differences in accident rates across categories are observed with segment length of 100m, with the significance gradually diminishing as segment length increases and disappearing entirely when the length exceeds 300m.

IMPLICATION

These findings validate the reliability of using alignment perspective features (APS and APK) for road alignment classification and road safety analysis, providing valuable insights for road safety management.

How Can the Trust-Change Direction be Measured and Identified During Takeover Transitions in Conditionally Automated Driving? Using Physiological Responses and Takeover-Related Factors

OBJECTIVE

This paper proposes an objective method to measure and identify trust-change directions during takeover transitions (TTs) in conditionally automated vehicles (AVs).

BACKGROUND

Takeover requests (TORs) will be recurring events in conditionally automated driving that could undermine trust, and then lead to inappropriate reliance on conditionally AVs, such as misuse and disuse.

METHOD

34 drivers engaged in the non-driving-related task were involved in a sequence of takeover events in a driving simulator. The relationships and effects between drivers' physiological responses, takeover-related factors, and trust-change directions during TTs were explored by the combination of an unsupervised learning algorithm and statistical analyses. Furthermore, different typical machine learning methods were applied to establish recognition models of trust-change directions during TTs based on takeover-related factors and physiological parameters.

RESULT

Combining the change values in the subjective trust rating and monitoring behavior before and after takeover can reliably measure trust-change directions during TTs. The statistical analysis results showed that physiological parameters (i.e., skin conductance and heart rate) during TTs are negatively linked with the trust-change directions. And drivers were more likely to increase trust during TTs when they were in longer TOR lead time, with more takeover frequencies, and dealing with the stationary vehicle scenario. More importantly, the F1-score of the random forest (RF) model is nearly 77.3%.

CONCLUSION

The features investigated and the RF model developed can identify trust-change directions during TTs accurately.

APPLICATION

Those findings can provide additional support for developing trust monitoring systems to mitigate both drivers' overtrust and undertrust in conditionally AVs.

Assessing driver behavior in work zones: A discretized duration approach to predict speeding

Higher speeds in work zones have been linked to an increased likelihood of crashes and more severe crash outcomes. To enhance safety, speed limits are often reduced in work zones, aiming to create a steady flow of traffic and safer traffic operations such as merging and flagging. However, this speed reduction can also lead to abrupt speed changes, resulting from sudden braking or acceleration, increasing the risk of crashes. This disruption in speed and flow results increases the likelihood of rear-end crashes. Ensuring driver compliance with the reduced speed limits and traffic flow operations is challenging as work zones may cause frustration and lead to more instances of speeding. Therefore, proactively predicting speeding events in work zones can be crucial for the safety of both workers and road users, as it enables the implementation of speed enforcement measures to maintain and improve driver compliance in advance. In this study, we employ the duration-based prediction framework to forecast speeding occurrences in work zones. The model is used to identify significant predictors of speeding including visibility, number of lanes, posted speed limit, segment length, coefficient of variation in speed, and travel time index. Among these variables, the number of lanes, posted speed limit, and coefficient of variation of speed are positively associated with speeding. On the other hand, visibility, segment length, and travel time index are negatively associated with speeding. Results show the model's predictive accuracy is higher for speeding events with shorter durations between consecutive occurrences. The model predicted speeding within 61% of the actual epoch when speeding events within 5 h of one another were considered for validation. This indicates that the model is more effective for road segments and work zones where speeding occurs more frequently. The prediction framework can be a great asset for agencies to improve work zone safety in real-time by enabling them to proactively implement effective work zone enforcement measures to control speeding and to stay prepared, preventing potential hazards.

Personality, perceptions and behavior: A study of speeding amongst drivers in Victoria, Australia

INTRODUCTION

Road crashes present a serious public health issue. Many people are seriously or fatally injured every year in avoidable crashes. While these crashes can have multiple contributing factors, including road design and condition, vehicle design and condition, the environment and human error, the performance of illegal driving behavior, including speeding, may also play a role. The current study aimed to examine the mediating influence that four potential deterrents (perceptions towards enforcement, crash risk, social norms and disapproval, and negative personal/emotional affect) have between the Big Five personality traits (conscientiousness; extraversion; agreeableness; neuroticism; openness) and expectations to speed.

METHODS

A total of 5,108 drivers in Victoria, Australia completed an online survey in 2019. A mediated regression analysis was used to examine pathways in a conceptual model developed for the study.

RESULTS

The results showed that perceptions towards the four potential deterrents examined did mediate the relationship (either completely or partially) between personality and expectations to speed.

CONCLUSIONS

The results of this study suggest that if interventions to deter illegal driving behavior are to be successful, one factor that could be taken into account is the personality traits of drivers who may be at greatest risk of the performance of illegal driving behaviors.

Application of naturalistic driving data: A systematic review and bibliometric analysis

The application of naturalistic driving data (NDD) has the potential to answer critical research questions in the area of driving behavior assessment, as well as the impact of exogenous and endogenous factors on driver safety. However, the presence of a large number of research domains and analysis foci makes a systematic review of NDD applications challenging in terms of information density and complexity. While previous research has focused on the execution of naturalistic driving studies and on specific analysis techniques, a multifaceted aggregation of NDD applications in Intelligent Transportation System (ITS) research is still unavailable. In spite of the current body of work being regularly updated with new findings, evolutionary nuances in this field remain relatively unknown. To address these deficits, the evolutionary trend of NDD applications was assessed using research performance analysis and science mapping. Subsequently, a systematic review was conducted using the keywords "naturalistic driving data" and "naturalistic driving study data". As a result, a set of 393 papers, Published between January 2002-March 2022, was thematically clustered based on the most common application areas utilizing NDD. the results highlighted the relationship between the most crucial research domains in ITS, where NDD had been incorporated, and application areas, modeling objectives, and analysis techniques involving naturalistic databases.

A novel collision warning system based on the visual road environment schema: An examination from vehicle and driver characteristics

Drivers pay unequal attention to different road environmental elements and visual fields, which greatly influences their driving behavior. However, existing collision warning systems ignore these visual characteristics of drivers, which limits the performance of collision warning systems. Therefore, this study proposes a novel collision warning system based on the visual road environment schema, in order to enhance the support for avoiding potential dangers in objects and areas that are easily overlooked by the drivers' vision. To capture the above visual characteristics of drivers, the visual road environment schema that consists of the semantic layer, the scene depth layer, the sensitive layer, and the visual field layer is established by using several different deep neural networks, which realizes the recognition, quantization, and analysis of the road environment from the drivers' visual perspective. The effectiveness of the novel collision warning system is verified by the driving simulation experiment from six indicators, including warning distance, maximum lateral acceleration, maximum longitudinal deceleration, minimum collision time, reaction time, and heart rate. Additionally, a grey target decision-making model is built to comprehensively evaluate the system. The results show that compared with the traditional collision warning system, the novel collision warning system proposed in this study performs significantly better and can discover potential dangers earlier, give timely warnings, enhance the vehicles' lateral stability and driving comfort, shorten reaction time, and relieve the drivers' nervousness. By integrating the drivers' visual characteristics into the collision warning system, this study could help to optimize the existing collision warning system and promote the mutual understanding between intelligent vehicles and human drivers.

An interpretable prediction model of illegal running into the opposite lane on curve sections of two-lane rural roads from drivers' visual perceptions

Illegal running into the opposite lane (IROL) on curve sections of two-lane rural roads is a frequently hazardous behavior and highly prone to fatal crashes. Although driving behaviors are always determined by the information from drivers' visual perceptions, current studies do not consider visual perceptions in predicting the occurrence of IROL. In addition, most machine learning methods belong to black-box algorithms and lack the interpretation of prediction results. Therefore, this study aims to propose an interpretable prediction model of IROL on curve sections of two-lane rural roads from drivers' visual perceptions. A new visual road environment model, consisting of five different visual layers, was established to better quantify drivers' visual perceptions by using deep neural networks. In this study, naturalistic driving data was collected on curve sections of typical two-lane rural roads in Tibet, China. There were 25 input variables extracted from the visual road environment, vehicle kinematics, and driver characteristics. Then, XGBoost (eXtreme Gradient Boosting) and SHAP (SHapley Additive exPlanation) methods were combined to build a prediction model. The results showed that our prediction model performed well, with an accuracy of 86.2% and an AUC value of 0.921. The average lead time of this prediction model was 4.4 s, sufficient for drivers to respond. Due to the advantages of SHAP, this study interpreted the impacting factors on this illegal behavior from three aspects, including relative importance, specific impacts, and variable dependency. After offering more quantitative information on the visual road environment, the findings of this study could improve the current prediction model and optimize road environment design, thereby reducing IROL on curve sections of two-lane rural roads.

Divergent Effects of Factors on Crash Severity under Autonomous and Conventional Driving Modes Using a Hierarchical Bayesian Approach

Influencing factors on crash severity involved with autonomous vehicles (AVs) have been paid increasing attention. However, there is a lack of comparative analyses of those factors between AVs and human-driven vehicles. To fill this research gap, the study aims to explore the divergent effects of factors on crash severity under autonomous and conventional (i.e., human-driven) driving modes. This study obtained 180 publicly available autonomous vehicle crash data, and 39 explanatory variables were extracted from three categories, including environment, roads, and vehicles. Then, a hierarchical Bayesian approach was applied to analyze the impacting factors on crash severity (i.e., injury or no injury) under both driving modes with considering unobserved heterogeneities. The results showed that some influencing factors affected both driving modes, but their degrees were different. For example, daily visitors' flowrate had a greater impact on the crash severity under the conventional driving mode. More influencing factors only had significant impacts on one of the driving modes. For example, in the autonomous driving mode, mixed land use increased the severity of crashes, while daytime had the opposite effects. This study could contribute to specifying more appropriate policies to reduce the crash severity of both autonomous and human-driven vehicles especially in mixed traffic conditions.

Utilization of Drivers' Dynamic Visual Characteristics to Find the Appropriate Information Quantity of Traffic Engineering Facilities on Straight Roads of Grassland Highways

To find the appropriate range of information quantity, we studied how the information quantity of traffic engineering facilities (TEFs) on straight roads of grassland highways affects a driver's eye movements. We used a combination of survey, statistics, analysis of variance, and the establishment of virtual scene to do this research, and carried out simulated driving tests at six levels (Z0, Z1, Z2, Z3, Z4, and Z5) of TEF information quantity. The driver's fixation duration, visual search breadth, and glance speed were evaluated in a quantitative way. Results showed that the information quantity had a significant impact on eye movements. It is concluded that the information quantity from 0 to 10 bits/km may cause problems to drivers, whereas the information quantity of 40 bits/km serves as the limit. The information quantity from 30 to 40 bits/km is the appropriate one for TEF on grassland highways.

A hybrid deep learning approach for driver anomalous lane changing identification

Reliable knowledge of driving states is of great importance to ensure road safety. Anomaly detection in driving behavior means recognizing anomalous driving states as a direct result of either environmental or psychological factors. This paper provides an efficient anomaly recognition approach to identify anomalous lane-changing events in a personalized manner. The proposed framework includes three unsupervised algorithms. First, a Recurrent-Convolutional Autoencoder extracts the spatio-temporal characteristics from a high-dimensional naturalistic driving dataset. Second, in order to recognize anomalous lane-changing events of individual drivers, the extracted latent feature space is analyzed using Pauta criterion-based reconstruction loss analysis, as well as one-class Support Vector Machine. Last, t-Distributed Stochastic Neighbor Embedding is employed to visualize the latent space for better understanding and interpretability. Temporal anomalies of lane-changing events were analyzed by a personalized grey relational coefficient analysis, to represent robust similarities for individual drivers. Validation and calibration were performed with a natural driving study dataset collected from 50 drivers with 59,372 lane change events. The results showed heterogeneity in the pattern of abnormal lane changing behavior across the sample. At the same time, each driver exhibited heterogeneous anomalous behaviors in both temporal and spatial sequences. Without prior labels, the proposed model effectively captures personalized driving patterns and abnormal lane-changing events from high-dimensional time-series data. This unsupervised hybrid approach is a novel attempt to complete personalized anomalous lane-changing behaviors identification based on naturalistic driving data involving various traffic environments. Our approach enables the extraction of natural individual lane-changing behavior patterns and provides insights for the improvement of personalized driving behavior monitoring systems.

Global lessons learned from naturalistic driving studies to advance traffic safety and operation research: A systematic review

The state of practice of investigating traffic safety and operation is primarily based on traditional data sources, such as spot sensors, loop detectors, and historical crash data. Recently, researchers have utilized transportation data from instrumented vehicles, driving simulators, and microsimulation modeling. However, these data sources might not represent the actual driving environment at a trajectory level and might introduce bias due to their experimental control. The shortcomings of these data sources can be overcome via Naturalistic Driving Studies (NDSs) considering the fact that NDS provides detailed real-time driving data that would help investigate the safety and operational impacts of human behavior along with other factors related to weather, traffic, and roadway geometry in a naturalistic setting. With the enormous potential of the NDS data, this study leveraged the Preferred Reporting Items for Systematic Reviews and Meta-Analyzes (PRISMA) approach to shortlist the most relevant naturalistic studies out of 2304 initial studies around the world with a focus on traffic safety and operation over the past fifteen years (2005-2020). A total of 117 studies were systematically reviewed, which were grouped into seven relevant topics, including driver behavior and performance, crash/near-crash causation, driver distraction, pedestrian/bicycle safety, intersection/traffic signal related studies, detection and prediction using NDSs data, based on their frequency of appearance in the keywords of these studies. The proper deployment of Connected and Autonomous Vehicles (CAV) require an appropriate level of human behavior integration, especially at the intimal stages where both CAV and human-driven vehicles will interact and share the same roadways in a mixed traffic environment. In order to integrate the heterogeneous nature of human behavior through behavior cloning approach, real-time trajectory-level NDS data is essential. The insights from this study revealed that NDSs could be effectively leveraged to perfect the behavior cloning to facilitate rapid and safe implementation of CAV.

Effects of an integrated collision warning system on risk compensation behavior: An examination under naturalistic driving conditions

Collision warning systems can improve traffic safety, while their safety benefit may be lessened due to improper risk compensation or system misuse. There are limited studies of advanced safety systems increasing unexpected risky driving behavior, especially with adolescent drivers. This study is designed to address this research gap in two main areas: 1) it seeks to examine whether and how the introduction of advanced driver-assistance systems influences drivers' risk compensation behavior (e.g., increase of hard braking frequency), and 2) it investigates key factors (e.g., distraction) that contribute to changes in hard braking frequency during driving for both teen and adult drivers. Naturalistic driving data from two previous studies were analyzed in this study with two methods: a hierarchical logistic regression model was used to evaluate the effects of an integrated collision warning system on hard braking behavior, while a Random forests algorithm was applied to model hard braking behavior and to rank the contributing factors by calculating the importance scores. No statistical evidence was observed that the integrated collision warning system significantly changed the likelihood of hard braking for teen or adult drivers. Other factors like distraction, especially visual-manual distraction, had the largest impact on the hard braking behavior, followed by speeding and roadway segments (i.e., at intersections or not). Short time-headways and driving in high-density traffic significantly increased the likelihood of hard braking. Furthermore, the rate of hard braking behavior on surface roads was much higher than on highways, as expected. Compared with straight road segments, hard braking behavior was less likely to occur on curve roads. This study applied an analytical strategy by using both machine learning and statistical analysis methods to achieve high model accuracy and facilitate inference concerning the relationships among variables. Findings in this study can help to improve the design of integrated collision warning systems and the use of autonomous braking systems, and to apply appropriate analysis methods in understanding teen drivers' behavior changes with those safety systems.

Operational factor analysis of the aggressive taxi speeders using random parameters Bayesian LASSO modeling approach

Partial taxi speeders are observed with both high speeding frequency and severity (range). They thereby can be viewed as aggressive speeders whose behaviors may result in more hazards than others. Among the factors contributing to taxi speeding, the operational factors are proven to be deterministic. However, previous studies mainly investigate the operational factors of taxi speeding frequency, which fail to comprehensively unveil the impact of factors on speeders, especially for aggressive speeders. This study intends to disclose the operational factors affecting the aggressive taxi speeders with the random parameters Bayesian least absolute shrinkage and selection operator (LASSO) modeling approach. Taxi speeding behaviors and several operational factors are extracted from taxi GPS trajectory data in Chengdu, China. Based on the hourly speeding frequency and average speeding severity of each speeder, the fuzzy C-means clustering algorithm is employed to categorize taxi speeders into three cohorts: restrained speeder (RS), moderate speeder (MS), and belligerent speeder (BS). Compared to RS, MS and BS are treated as the aggressive taxi speeders. Several binary logistic models are developed with RS as the reference category. The random parameters Bayesian binary logistic LASSO model that captures the unobserved heterogeneity and tackles the multicollinearity is found to be the best fit model to identify the significant operational factors. The results indicate that aggressive taxi speeders are linked to longer daily driving distance and cruise distance, shorter delivery time, higher hourly income, driving at night, and driving on low-speed limit roads. However, intensive lane-changes and sufficient daily naps do not contribute to aggressive taxi speeders. Moreover, BS is more sensitive to the operational factors than MS. This study stresses the necessity of implementing speeder classification in taxi driver management and conceiving countermeasures considering the operational factors which are significantly associated with the aggressive taxi speeders.

Is organized carpooling safer? Speeding and distracted driving behaviors from a naturalistic driving study in Brazil

Carpooling consists of drivers and passengers sharing a journey and its costs. Nowadays, in the context of mobility as a service, organized carpooling encompasses a service and trust relationship between drivers and passengers, by matching common routes and splitting cost through mobile phone applications. Therefore, passengers expect a certain level of travel quality and safety. In this context, this research aims to verify the hypothesis that drivers in an organized carpooling situation (CP) show safer driving behavior in terms of speeding (SP) and mobile phone use while driving (MPU) in comparison with non-carpooling (NCP) drivers. The research is based on data from the Brazilian Naturalistic Driving Study (NDS-BR) conducted in the City of Curitiba, with 40.45 driving hours and a traveled distance of 895.87 km. Methodology included the selection of safety performance indicators on SP and MPU, use of nonparametric Wilcoxon signed rank test for safety performance indicator comparisons and Pearson Chi-Square to test the association between CP or NCP and low or high indicator values. Hypothesis test results point in the same direction and partially confirm the initial assumption that CP induces safer behavior in terms of speeding. The statistically sound results showed that CP drivers engaged in less speeding episodes and mobile phone use duration in comparison to NCP drivers, as well as lower speed while using a mobile phone. In addition, driver behavior in CP and NCP situations also differed in terms of the type of MPU, with the proportion of types of use that demand a higher level of visual and manual distraction being higher among NCP drivers. In summary, these results confirm the initial hypothesis of safer driving behavior during carpooling in terms of MPU while driving.

Analyzing driver behavior under naturalistic driving conditions: A review

For a decade, researchers working in the area of road safety have started exploring the use of driving behavior data for a better understanding of the causes related to road accidents. A review of the literature reveals the excellent potential of naturalistic driving studies carried out by collecting vehicle performance data and driver behavior data during normal, impaired, and safety-critical situations. An in-depth understanding of driver behavior helps analyze and implement pre-crash safety measures - the development of enforcement policies, infrastructure design, and intelligent vehicle safety systems. The present paper attempts to review the naturalistic driving studies that have been undertaken so far. The paper begins with an overview of different methods for collecting unobtrusive driver behavior data during their day to day trip, followed by a discussion of various factors affecting driving behavior and their influence on vehicle performance parameters. The paper also discusses the strategies mentioned in the literature for improving driving behavior using naturalistic driving studies to enhance road safety. Some of the major findings of this review suggest that i) driver behavior is a major cause in the majority of the road accidents ii) drivers generally reduce their speed and increases headway as a compensatory measure to reduce the workload imposed during distracting activity and adverse weather conditions iii) mobile phone has emerged as a potential device for collecting naturalistic driving data and, iv) improvement in driving behavior can be achieved by providing feedback to the drivers about their driving behavior. This can be done by implementing usage-based insurance schemes such as pay as you drive (PAYD), pay how you drive (PHYD), and manage how you drive (MHYD). While a considerable amount of research has been done to analyze driving behavior under naturalistic conditions, some areas which are yet to be explored are highlighted in the present paper.

Quantifying the Effects of Visual Road Information on Drivers' Speed Choices to Promote Self-Explaining Roads

Roads should deliver appropriate information to drivers and thus induce safer driving behavior. This concept is also known as “self-explaining roads” (SERs). Previous studies have demonstrated that understanding how road characteristics affect drivers’ speed choices is the key to SERs. Thus, in order to reduce traffic casualties via engineering methods, this study aimed to establish a speed decision model based on visual road information and to propose an innovative method of SER design. It was assumed that driving speed is determined by road geometry and modified by the environment. Lane fitting and image semantic segmentation techniques were used to extract road features. Field experiments were conducted in Tibet, China, and 1375 typical road scenarios were picked out. By controlling variables, the driving speed stimulated by each piece of information was evaluated. Prediction models for geometry-determined speed and environment-modified speed were built using the random forest algorithm and convolutional neural network. Results showed that the curvature of the right boundary in “near scene” and “middle scene”, and the density of roadside greenery and residences play an important role in regulating driving speed. The findings of this research could provide qualitative and quantitative suggestions for the optimization of road design that would guide drivers to choose more reasonable driving speeds.