Measuring geometric properties for large-sized parts is faced with various challenges. This study established a novel method for measuring the flatness of large-scale annular workpieces, with the flange of a wind turbine tower as an example. The proposed method is based on three-dimensional (3D) machine vision and precision motion control technology, aiming to address issues such as distorted measurements, long measurement times, and significant safety risks. The study primarily focused on the development of algorithms for measuring the flatness of large-sized annular workpieces through region-based measurement and vertical centering. The system employs a rotating arm equipped with a 3D vision sensor driven by a precision motion control system. A constant-speed laser line scans the surface of the large-sized flange to acquire precise point cloud data. The obtained point cloud data are then processed to determine the flatness of the flange surface. A verification test was conducted on a wind turbine tower flange with a diameter of 4.2 m to validate the functionality of the system. The results indicated that the proposed system enables stable and accurate measurement of flange flatness, and the measurement uncertainty was analyzed using the guide to the uncertainty in the measurement method, yielding an expanded uncertainty of 0.072 mm. The results indicated that the developed flatness measurement system can efficiently, accurately, and consistently measure the flatness of large annular flanges. This provides valuable insights for the geometric measurement of large annular parts in various fields such as wind power, engineering machinery, ships, and military industries.

To address the need for rapid acquisition of package volume in the logistics industry, a method for measuring the volume of logistics packages based on speckle structured light cameras is proposed. This method utilizes an Astra structured light camera to construct a virtual plane. The plane is obtained by fitting a point cloud on the workbench, and a plane correction algorithm is designed to achieve plane correction and obtain a reference plane. The measurement position of the package is determined through depth image processing. An algorithm for projecting the edge point cloud onto the reference plane is proposed, combined with the minimum bounding rectangle algorithm to measure the length and width of the package. The height of the package is measured using a method of precision error compensation. Through experimental verification and uncertainty analysis, this method enables rapid and stable measurement of the volume of regular logistics packages. Additionally, it is also applicable for measuring irregular logistics packages. According to the analysis based on the guide to the uncertainty in the measurement method, the expanded uncertainties of the measured length and width are 3.57 mm, and the expanded uncertainty of the measured height is 2.20 mm, meeting the requirements for parcel volume measurement. This method demonstrates a stable and reliable volume measurement effect, enabling efficient and accurate acquisition of package volume measurement results.