A Survey: Industrial Anomaly Detection based on Data Mining

Industrial defect detection plays a crucial role in modern manufacturing. Identifying and addressing inferior products contributes to enhancing product quality, strengthening product competitiveness, and increasing customer satisfaction. Existing surveys of industrial defect detection are relatively scarce and struggle to reflect the latest development trends. Therefore, this article provides a more detailed and in-depth survey of industrial defect detection technologies. The article first reviews the development history of industrial defect detection methods. It then covers three aspects: the concept of general anomalies, concepts related to image anomaly detection, and industrial defects, providing an overview of industrial defect detection in these areas. It also summarizes the current state of development, as well as the advantages and disadvantages of each aspect. Additionally, the article identifies the limitations of industrial detection methods in practical industrial applications. Finally, it looks forward to the future development trends and potential research directions in this field, aiming to inspire future research.