During the LPBF (Laser Powder Bed Fusion) process, excessive residual stress can lead to a decline in the forming quality of printed parts. Residual stress is the primary cause of part deformation; therefore, monitoring the part deformation trend can, to a certain extent, reflect the overall residual stress level of the part. Online stress or deformation monitoring presents significant challenges due to the part being covered by powder and the need for a sealed environment. This paper proposes a novel monitoring system that replaces the bolts used to fix the baseplate with bolts capable of measuring tensile force. The deformation trend of the part will be inhibited by the bolts fixing it to the base plate, and the magnitude of the bolt tensile force is indicative of the magnitude of the deformation trend. Printing experiments were conducted under different scanning strategies and preheating conditions. The experimental curves reveal the patterns of deformation trend evolution in the parts, and these patterns are well explained by existing theories, which confirms the reliability and accuracy of the monitoring system.

Visual cryptography is a type of image encryption technology with higher levels of security, concealment, and ease of use than traditional encryption algorithms. Therefore, visual cryptography can be combined with digital watermarking technology to enhance the security of copyright protection processes. Currently, all types of digital watermarking schemes directly embed the copyrighted image into the carrier image, which cannot avoid passive attacks from malicious extraction. This paper proposes a copyright protection scheme for QR codes under unextended visual cryptography to address the issue of copyright protection. On the one hand, using the shadow image generated by visual cryptography for watermark embedding, even if maliciously extracted, will not disclose any content of the copyrighted image; On the other hand, a third-party copyright authentication center is introduced to authenticate the identity of the copyright owner, which enhances the security of copyright image restoration. Finally, the proposed scheme divides the QR code that displays necessary information into blocks and applies discrete cosine transform. The watermark is embedded into the feature value matrix obtained from the singular value decomposition of the central block, ensuring both robustness and invisibility of the watermark while avoiding malicious removal. After experimental results, it was shown that using the watermark embedding algorithm proposed in this paper for copyright image embedding, the copyright image has excellent robustness, invisibility, and higher security.