Detecting Defects in Manufacturing Process Using Deep Learning Models

The industrial business needs to detect casting faults. In this research, we provide a deep learning-based technique for detecting casting errors. We offer the Efficient network with depth separable convolution to reduce computer resources while retaining detection accuracy in order to address the issue of limited computing resources that plagues the majority of manufacturing firms. The experiment's findings showthat our strategy can still work well even when faced with limited processing power. It offers an automated casting fault detection technique and makes a suggestion to researchers looking into industrial problems who were having trouble getting enough computational power. Defect detection is often part of discretely made components quality control. Activities that need lengthy, repetitive work must be done by employees with extensive training and/or expertise.