Pioneering Driver Safety: Evaluating Weather Impacts with the Multi-Class-Weather Algorithm

Ensuring road safety requires a detailed understanding of traffic scene perception and driver risk assessment. This study focuses on the impact of weather conditions on driver injury severity, using a novel algorithm called Multi-Class-Weather, which integrates the advantages of ResNet9, multinomial logic, and self-learning. The relationship between different weather conditions and driver injury levels was carefully studied using deep supervised learning methods. A multinomial logistic regression model was used to quantitatively measure this relationship. Experimental results show a positive correlation between foggy weather conditions and injury severity. At the same time, after comparison of experiments, it was found that the new algorithm is helpful in preventing model overfitting, and the accuracy rate reaches 87%. The results of this study provide important insights into driver risk assessment under different weather conditions and highlight the effectiveness of multi-class weather algorithms in this specialized field.