DMMCNet: Dynamic Multiscale and Multilevel Contrastive Point Cloud Feature Extraction Network for Industrial Defect Detection

Industrial defect detection is crucial for ensuring product quality and safety. Point-cloud-based methods have attracted increasing attention due to their accurate geometric feature extraction capability. However, most of them first, lack the ability to dynamically adjust multiscale receptive fields, limiting their capacity to perceive defects with diverse sizes, second, ignore semantic inconsistencies among multiscale features during fusion, weakening the ability to distinguish defect regions, and third, pay insufficient attention to key points, lacking sensitivity to fine-grained defects. To address these issues, we propose a dynamic multiscale and multilevel contrastive point cloud feature extraction network (DMMCNet) for industrial defect detection, which consists of three key modules: dynamic multiscale feature encoding module, multilevel contrastive fusion module, and key-point enhancement module (KPE). Specifically, in DMFE, we propose a learnable scale embedding mechanism that dynamically adjusts receptive field sizes according to the geometric context of the point cloud, improving the perception of defects with diverse sizes. In MCF, we develop a multilevel fusion scheme that progressively aggregates multiscale features, allowing fine-scale defect cues to be preserved before incorporating background information. Within this fusion process, we propose a cross-scale contrastive constraint mechanism to align features from different receptive-field scales and alleviate semantic inconsistencies. In KPE, we model spatial correlations among points to guide the network’s attention toward key points with distinct geometric variations, strengthening the response to critical defect areas. Finally, experimental results on three industrial point cloud datasets demonstrate that DMMCNet outperforms thirteen state-of-the-art methods.