Highlights Inception-v3 deep learning model has been developed to predict pedestrian-involved crash severity using data collected over five years (2016–2021) from Louisiana. The Boruta algorithm was applied to determine the importance of variables and investigate the contributing factors to pedestrian-involved crash severity. Factors associated with pedestrian-involved crash severity include pedestrian gender, pedestrian and driver impairment, posted speed limits, alcohol involvement, pedestrian age, visibility obstruction, roadway lighting conditions, and both pedestrian and driver conditions, including distraction and inattentiveness. Deep learning models with data balancing using both over and under sampling can achieve superior prediction performance compared to models without data balancing. Comparative analysis and McNemar’s tests revealed the superiority of the Inception-v3 deep learning model over statistical and conventional machine learning models.
Abstract This research leverages a novel deep learning model, Inception-v3, to predict pedestrian crash severity using data collected over five years (2016–2021) from Louisiana. The final dataset incorporates forty different variables related to pedestrian attributes, environmental conditions, and vehicular specifics. Crash severity was classified into three categories: fatal, injury, and no injury. The Boruta algorithm was applied to determine the importance of variables and investigate contributing factors to pedestrian crash severity, revealing several associated aspects, including pedestrian gender, pedestrian and driver impairment, posted speed limits, alcohol involvement, pedestrian age, visibility obstruction, roadway lighting conditions, and both pedestrian and driver conditions, including distraction and inattentiveness. To address data imbalance, the study employed Random Under Sampling (RUS) and the Synthetic Minority Oversampling Technique (SMOTE). The DeepInsight technique transformed numeric data into images. Subsequently, five crash severity prediction models were developed with Inception-v3, considering various scenarios, including original, under-sampled, over-sampled, a combination of under and over-sampled data, and the top twenty-five important variables. Results indicated that the model applying both over and under sampling outperforms models based on other data balancing techniques in terms of several performance metrics, including accuracy, sensitivity, precision, specificity, false negative ratio (FNR), false positive ratio (FPR), and F1-score. This model achieved prediction accuracies of 93.5%, 77.5%, and 85.9% for fatal, injury, and no injury categories, respectively. Additionally, comparative analysis based on several performance metrics and McNemar’s tests demonstrated that the predictive performance of the Inception-v3 deep learning model is statistically superior compared to traditional machine learning and statistical models. The insights from this research can be effectively harnessed by safety professionals, emergency service providers, traffic management centers, and vehicle manufacturers to enhance their safety measures and applications.
Predicting pedestrian-involved crash severity using inception-v3 deep learning model
2024-01-02
Article (Journal)
Electronic Resource
English
Analysis of vehicle pedestrian crash severity using advanced machine learning techniques
DOAJ | 2023
|A Multinomial Logit Model of Pedestrian-Vehicle Crash Severity
Online Contents | 2011
|A Multinomial Logit Model of Pedestrian-Vehicle Crash Severity
Taylor & Francis Verlag | 2011
|Modelling pedestrian crash severity at mid-blocks
Taylor & Francis Verlag | 2017
|High alcohol-involved pedestrian crash sites: driver and pedestrian behaviors
British Library Conference Proceedings | 1997
|