Developing A Machine Learning-Based Computer Vision Model for External Defect Detection in Camera Components

This is the authors' abstract. We don't add key points for this paper.

In this article, through exploring the operational procedures of domestic component manufacturers, specifically the production process of cameras for technological products, it is evident that post-processing product defect control poses a significant challenge. This process requires intervention from experienced personnel to conduct manual inspections, which has led to decreased productivity and diminished reliability. Based on this scenario, this paper proposes a computer vision model to detect visual defects in camera components. Machine learning and deep learning will be the primary technologies applied to overcome this challenge. The proposed method is evaluated in terms of accuracy and high performance. However, the model still requires further architectural adjustments and increased training data to achieve optimal refinement. The gained results will significantly enhance the effective of industrial camera production.