Integrating Deep Learning for Reliable Tyre Defect Classification in Automated Systems

This work researches the classification of tyres as good and defective using deep learning models, which are ResNet50 and VGG19. This could be due to early tyre trouble detection as a means of the success of vehicle safety and assured machine learning can greatly enhance precision and effectiveness. As compared to ResNet50, VGG19 attained 88.98% accuracy. The VGG19 model has a relatively simple design that makes it a better performer compared to ResNet50, especially when dealing with smaller datasets and relatively easy classification of images. The new insights that the study brings into the utilisation of deep learning for tyre classification pave the way for developing better vehicle safety systems and smart manufacturing.