Tire Condition Classification Based on Tread Depth using Machine Learning

As a crucial component for vehicle performance and stability, automotive manufacturers adopt specific tire inspection procedures to verify its quality and integrity. Tires stay in good condition when they are checked frequently for things like tread depth and pressure. Moreover, tire tread depth has a significant impact on a vehicle's braking ability and should be routinely evaluated to ensure driver safety. In order to identify and classify tire condition, a state-of-the-art research approach is carried out using machine learning algorithms: Extreme Gradient Boosting (XGBoost), AdaBoost, Naïve Bayes, Support Vector Machine (SVM), Decision Tree and Random Forest. A primary data set is taken from tire tread depth measurements on 100 tires using tread depth gauges and tire pressure gauges. The results recorded 95% accuracy and recall, 95.35% F1 and 96.67% precision scores for both XGBoost and SVM. The outcome shows the successful implementation of machine learning algorithms for tire condition classification.