Advance in 3D printing defect detection technology based on deep learning

• Published in: 2023 IEEE 3rd International Conference on Power, Electronics and Computer Applications (ICPECA) pp. 413 - 418
• Main Authors: Su, Yizhou, Guan, Huaiguang, Wang, Xunwei, Qi, Guanghao, Lei, Baozhen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 29.01.2023
• Subjects: 3D printing; Additive manufacturing; Deep learning; Defect detection; Electric potential; Manufacturing industries; Nondestructive testing; Production; Three-dimensional displays; Three-dimensional printing
• DOI: 10.1109/ICPECA56706.2023.10075837

With the development of manufacturing industry, defect detection is more and more widely used in industrial production, and there are more and more strict requirements for the accuracy and efficiency of detection. Compared with traditional manufacturing methods, 3D printing, with its unique manufacturing method, has unique advantages in small batch and multi batch manufacturing. This review aims to introduce some of the mainstream methods of industrial defect detection. First, it introduces the background and characteristics of industrial defect detection, introduces the industrial CT of classical nondestructive testing methods, and introduces the classical methods of defect detection and deep learning methods. The 3D reconstruction of defects is also introduced, and the methods of obtaining defect detection data and optimizing industrial parameters are introduced, and the commonly used defect detection data sets are summarized. Finally, the future development trend and potential research directions in this field are prospected.