Supervised Learning Plastic Defect Algorithm Detection

Abstract

The goal of this research is to develop a supervised learning algorithm able to detect the defects of plastic’s material. Finding patterns or examples in a dataset that differ from the norm is known as anomaly detection in plastic textures. Anomalies, in the context of plastic textures, can refer to imperfections’ deviations, or anomalies in the material that may have an impact on the final product's overall quality. Conventional anomaly detection techniques frequently rely on rule-based systems or manual examination, which can be laborious, subjective, and unable to identify small anomalies.