Vat photopolymerization is renowned for its high flexibility, efficiency, and precision in ceramic additive manufacturing. However, due to the impact of random defects during the recoating process, ensuring the yield of finished products is challenging. At present, the industry mainly relies on manual visual inspection to detect defects; this is an inefficient method. To address this limitation, this paper presents a method for ceramic vat photopolymerization defect detection based on a deep learning framework. The framework innovatively adopts a dual-branch object detection approach, where one branch utilizes a fully convolution network to extract the features from fused images and the other branch employs a differential Siamese network to extract the differential information between two consecutive layer images. Through the design of the dual branches, the decoupling of image feature layers and image spatial attention weights is achieved, thereby alleviating the impact of a few abnormal points on training results and playing a crucial role in stabilizing the training process, which is suitable for training on small-scale datasets. Comparative experiments are implemented and the results show that using a Resnet50 backbone for feature extraction and a HED network for the differential Siamese network module yields the best detection performance, with an obtained F1 score of 0.89. Additionally, as a single-stage defect object detector, the model achieves a detection frame rate of 54.01 frames per second, which meets the real-time detection requirements. By monitoring the recoating process in real-time, the manufacturing fluency of industrial equipment can be effectively enhanced, contributing to the improvement of the yield of ceramic additive manufacturing products.

Vat photopolymerization is characterized by its high precision and efficiency, making it a highly promising technique in ceramic additive manufacturing. However, the process faces a significant challenge in the form of recoating defects, necessitating real-time monitoring to maintain process stability. This article presents a defect detection method that leverages multi-image fusion and deep learning for identifying recoating defects in ceramic additive manufacturing. In the image fusion process, multiple single-channel recoating images captured by monitoring camera positioned near the photopolymerization equipment are merged with curing area mask image to create a three-channel color image. The recoating images suffer from perspective distortion due to their side view. To facilitate fusion with the curing area image, image rectification technique is applied to correct the perspective distortion, transforming the side view recoating images into a top-down view. Subsequently, the fused images are processed using a channel-wise YOLO (You Only Look Once, CW-YOLO) method to extract features, enabling the distinction of various types of defects. When compared with other deep learning models, CW-YOLO achieves higher detection accuracy while maintaining a detection rate of 103.58fps, meeting the requirements for real-time detection. Furthermore, the paper introduces the F1 score as a comprehensive evaluation metric, capturing both detection accuracy and recall rate. The results show that the F1 score is enhanced by approximately 10% after image fusion, demonstrating that the proposed method can significantly improve defect detection, particularly in cases involving difficult-to-distinguish defects like material shortages and scratches.

Point cloud models of industrial parts are characterized by an unordered distribution of points and locally similar structures. Traditional point cloud classification methods have limited capability in modeling local neighborhood relations and are insufficient in representing local geometric differences, which restricts their performance in industrial part classification tasks. To address this issue, a point cloud classification method fusing vectorized local relation modeling and geometric enhancement is proposed. Specifically, a Vector-Edge Local Aggregation Block is designed to reconstruct the conventional local feature extraction unit. First, vectorized local relation modeling is introduced to construct hybrid local descriptors, generate vector features, and perform aggregation as well as scalar projection, thereby explicitly encoding the geometric and structural relations between the centroid point and its neighboring points. Then, edge features are extracted to supplement local differential geometric information. Finally, residual fusion is employed to achieve collaborative representation of multiple features, enabling the network to jointly capture local relational patterns and differential geometric information, thus improving its ability to represent complex local structures and fine-grained inter-class differences. Experiments on ModelNet40, the industrial part dataset MCB, and S3DIS demonstrate that the proposed method improves local feature representation for point clouds, effectively distinguishes fine-grained structural differences among industrial parts, and achieves strong performance, robustness, and generalization ability in point cloud tasks.