PointSGRADE: Sparse learning with graph representation for anomaly detection by using unstructured 3D point cloud data

Abstract

Surface anomaly detection by using 3D point cloud data has recently received significant attention. To completely measure the common free-form surfaces without loss of details, advanced 3D scanning technologies, such as 3D laser scanners, can be applied and will produce an unstructured point cloud. However, this irregular data structure poses challenges to anomaly detection, in that the existing methods based on regular data, e.g., 2D image, cannot be directly applied. This article proposes a sparse learning framework with a graph representation of the unstructured point cloud for anomaly detection (PointSGRADE). Specifically, the free-form surface is assumed to be smooth. Then, the associated point cloud can be represented as a graph. Subsequently, considering the sparse anomalies, we propose a sparse learning framework and formulate the anomaly detection problem as a penalized optimization problem, which is further solved by a computationally efficient majorization-minimization framework. Case studies demonstrate the accuracy and robustness of the proposed method. This article proposes a novel methodology for sparse anomaly detection on smooth free-form surfaces represented by unstructured point cloud, which is critical for quality inspection in manufacturing and other application areas.

Keywords:

• Anomaly detection
• unstructured 3D point cloud data
• graph representation
• sparse learning
• majorization-minimization
• free-form surface
• smooth surface

Additional information

Funding

This work was supported by the National Natural Science Foundation of China under Grant No. 72001139 and No. 72371219, Guangdong Basic and Applied Basic Research Foundation under Grant No. 2023A1515011656, Guangzhou Municipal Science and Technology Program under Grant No. 202201011235, and Guangzhou-HKUST(GZ) Joint Funding Program under Grant No. 2023A03J0651.

Notes on contributors

Chengyu Tao

Chengyu Tao is pursuing his PhD degree in individualized interdisciplinary program (smart manufacturing) with the Division of Emerging Interdisciplinary Areas, The Hong Kong University of Science and Technology, Hong Kong SAR, China. He is also affiliated with Guangzhou HKUST Fok Ying Tung Research Institute, Guangzhou, China. His research interests focus on quality control for smart manufacturing systems by using data analytics and optimization tools.

Juan Du

Juan Du is currently an assistant professor with the Smart Manufacturing Thrust, Systems Hub, The Hong Kong University of Science and Technology (Guangzhou), China. She is also affiliated with the Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology, Hong Kong SAR, China, and Guangzhou HKUST Fok Ying Tung Research Institute, Guangzhou, China. Her current research interests include data analytics and machine learning for modeling, monitoring, control, diagnosis and optimization in smart manufacturing systems. Dr. Du is a senior member of the Institute of Industrial and Systems Engineers (IISE), and a member of the Institute for Operations Research and the Management Sciences (INFORMS), the American Society of Mechanical Engineers (ASME), and the Society of Manufacturing Engineers (SME).