Five-axis continuous scanning inspection is an emerging technology in measuring free-form surfaces. Compared to the traditional three-axis scanning inspection, the five-axis continuous surface scanning tremendously boosts the inspection efficiency and suffers from less dynamic error due to adopting a special sweep scanning working mode, which lets the super-light and high-speed rotary two-axis probe head take over the majority of the inspection work. However, there has not been any efficient method to generate sweep scanning path for a free-form surface with external obstacle. The major problem is how to guarantee the non-interference condition during the scanning process. In this paper, a two-step algorithm for generating smooth and interference-free sweep scanning path is proposed. First, an image-based algorithm is proposed to incrementally calculate the admissible area of the stylus along the scanning path. Then an optimized B-spline fitting algorithm is proposed to find the optimal probe head trajectory within the admissible area, which simultaneously guarantees the smoothness and non-interference of the trajectory. In the real experiment, the average tangential velocity and acceleration of the machine's translational axes with the proposed five-axis sweep scanning path are only 16.45% and 17.19% of those with three-axis zigzag scanning path respectively. When inspecting a high-accuracy cylindrical surface with high scanning velocity, the root mean square of the surface profile error is 0.002 mm and 0.004 mm in five-axis scanning and three-axis scanning respectively. The experimental result manifests that the sweep scanning path tremendously helps to reduce the kinematic loads and the dynamic errors of the inspection machine.

An image-based algorithm for generating smooth and interference-free five-axis sweep scanning path

Corners are an important kind of image feature and play a crucial role in solving various tasks. Over the past few decades, a great number of corner detectors have been proposed. However, there is no benchmark dataset with labeled ground-truth corners and unified metrics to evaluate their corner detection performance. In this paper, we build three benchmark datasets for corner detection. The first two consist of those binary and gray-value images that have been commonly used in previous corner detection studies. The third one contains a set of urban images, called the Urban-Corner dataset. For each test image in these three datasets, the ground-truth corners are manually labeled as objectively as possible with the assistance of a line segment detector. Then, a set of benchmark evaluation metrics is suggested, including five conventional ones: the precision, the recall, the arithmetic mean of precision and recall (APR), the F score, the localization error (Le), and a new one proposed in this work called the repeatability referenced to ground truth (RGT). Finally, a comprehensive evaluation of current state-of-the-art corner detectors is conducted.

/pdf/a-benchmark-for-the-evaluation-of-corner-detectors-2szkz7oe.pdf

A Benchmark for the Evaluation of Corner Detectors

Five-axis sweep scanning is an emerging technology for freeform surface inspection. This new inspection manner greatly improves measurement efficiency and reduces the dynamic errors. However, the existing five-axis scanning path planning methods ignore the obstacle avoidance issue during sweep scanning, which may cause damage to the probe stylus if any collision occurs. In this paper, the collision detection is performed by a graphics engine HOOPS. An incremental algorithm is proposed to reduce the computation of the collision detection. A collision-free sweep scanning path for a freeform surface patch can be calculated within 30 s. In real experiments, the tangential velocity and acceleration in the proposed five-axis sweep scanning path are only 39.3% and 20.1% of those in the traditional three-axis scanning, which manifests that the sweep scanning path helps to reduce the kinematic loads and the dynamic error of the inspection machine.

Generation of Collision-Free Five-Axis Sweep Scanning Path Using Graphics Engine

Tea flow rate is a key indicator in tea production and processing. Due to the small real−time flow of tea leaves on the production line, the noise caused by the transmission system is greater than or close to the real signal of tea leaves. This issue may affect the dynamic measurement accuracy of tea flow. Therefore, a variational mode decomposition combined with a wavelet threshold (VMD−WT) denoising method is proposed to improve the accuracy of tea flow measurement. The denoising method of the tea flow signal based on VMD−WT is established, and the results are compared with WT, VMD, empirical mode decomposition (EMD), and empirical mode decomposition combined with wavelet threshold (EMD−WT). In addition, the dynamic measurement of different tea flow in tea processing is carried out. The result shows that the main noise of tea flow measurement comes from mechanical vibration. The VMD−WT method can effectively remove the noise in the tea dynamic weighing signal, and the denoising performance is better than WT, VMD, EMD, and EMD−WT methods. The average cumulative measurement accuracy of the tea flow signal based on the VMD−WT algorithm is 0.88%, which is 55% higher than that before denoising. This study provides an effective method for dynamic and accurate measurement of tea flow and offers technical support for digital control of the tea processing.

/pdf/research-on-dynamic-measurement-method-of-flow-rate-in-tea-1qbbjo75.pdf

Research on Dynamic Measurement Method of Flow Rate in Tea Processing

A theoretical model for research the accuracy of a method for measuring deviations from flatness is considered in the publication. The measurement method is based on the concept of the Kalman filter and is particularly effective against the action of various types of external influences, such as vibrations and shocks, noises of electromagnetic origin, temperature deformations, etc. It is adapted to measurements in static mode. Based on the measurement method, specialized algorithms can be developed to automatically correct measurement errors of shape and placement deviations, which can significantly improve accuracy characteristics, simplify the construction of measuring instruments, and reduce metrological requirements for their components and elements. Based on the structure of the measurement method, the concept of the research model is developed, which is based on the mutual conditioning of the statistical characteristics of the errors of model and measurement, derived as standalone components in the base system of equations.

Accuracy Evaluation of Flat Surfaces Measurements in Conditions of External Influences

The best way to improve the accuracy of coordinate measuring machine (CMM) is to establish a correlation dynamic error correction model. However, due to the complex source of dynamic measurement error, the traditional method cannot work well. In this paper, decision regression tree is used to analyze the complex relationship between dynamic measurement error and its influencing factors, and the measurement error is predicted. The results show that the algorithm is simple, fast and superior to the least square estimation.

Analysis of CMM Dynamic Measurement Error Based on Decision Regression Tree

To improve the robustness and accuracy of the corner-detection algorithm, this paper proposes a camera-calibration method based on the EDLines algorithm for the automatic detection of chessboard corners. The EDLines algorithm is initially used to perform straight-line detection on the calibration image. The features of the broken straight lines at the corners are then used to filter the straight lines and remove the background straight lines outside the chessboard. The pixels in the rectangular area around the filtered straight line are sorted by the gray gradient. After using the sorted results to fit the straight line, the coordinates of the intersection of the straight lines are taken as the initial coordinates of the corners and perform subpixel optimization on them. Finally, the corner points are sorted by the conversion between pixel-coordinate systems. The camera exposure time changes and complex imaging-background experiments show that the algorithm has no missed detection and redundancy in corner detection. The average reprojection error is found to be less than 0.05 pixels, which can be used in actual calibration.

/pdf/chessboard-corner-detection-based-on-edlines-algorithm-2jcizcpu.pdf

Chessboard Corner Detection Based on EDLines Algorithm

Fish is a common and widely distributed creature. Its skin has a unique physiological structure and plays an important role in many fields. Fish skin also has important potential value for bionics research. This study aims to provide a method and a reliable data for the study of bionics. A method of measuring the mechanical properties of fish skin samples using a binocular stereo digital image correlation (DIC) system combined with a synchronous tensile testing machine was proposed. The mechanical properties (e.g., elastic modulus E and strain) of grass fish skin samples (GFSA) were tested in hydrophilic and dry states. A dual-frequency laser interferometer was used to calibrate the tensile testing machine synchronously, and the feasibility and strain accuracy of DIC in GFSA measurement were verified by finite element method (FEM). The results show differences in the mechanical properties of GFSA between different individuals, different parts, and different states. Under the same stress, the head was easy to deform, and the strain was the largest, and E was the smallest. The tail result was the opposite of the head result.

/pdf/mechanical-property-test-of-grass-carp-skin-material-based-3mz09dra.pdf

Mechanical Property Test of Grass Carp Skin Material Based on the Digital Image Correlation Method

The sources of dynamic measurement error of CMM are complex and influence each other and the traditional parameter modeling method is very difficult to model. In this paper, we propose a hybrid model which combines neural network quantile regression and kernel density estimation. The hybrid model realizes the advantages of multi-angle analysis, nonlinear fitting data and non-parametric error prediction. We use this model to analyze the complex relationship between dynamic measurement error value and 3d coordinates, positioning velocity, proximity distance and contact velocity. The results show that our model has good predictive performance and is superior to the least squares estimation model.

/pdf/error-prediction-of-cmm-using-a-hybrid-model-based-on-neural-48s90a29fd.pdf

Error Prediction of CMM Using a Hybrid Model Based on Neural Network Quantile Regression and Kernel Density Estimation

Abstract. We propose an ex-situ reposition error compensation method to solve the reposition error in the ex-situ state of 2D digital image correlation (2D-DIC) in the long-term deformation of the measurement specimen or under special load deformation. Using the dual-reflector auxiliary measurement system with an included angle of 90 deg, the two sides of the repositioned specimen are mapped into left and right mirrors. The in- and out-of-plane rigid body displacements are compensated via the error compensation method, and the real in-plane displacement of the tested specimen is obtained. The accuracy and practicability of this method are verified by conducting an in-plane rigid body displacement experiment, unloaded specimen reposition experiment, thermal deformation experiment, and finite element analysis. We provide a simple, effective, and high-precision deformation measurement method for 2D-DIC to realize ex-situ measurement.

Mei Zhang

Papers

Analysis of CMM Dynamic Measurement Error Based on Decision Regression Tree

Chessboard Corner Detection Based on EDLines Algorithm

Mechanical Property Test of Grass Carp Skin Material Based on the Digital Image Correlation Method

Error Prediction of CMM Using a Hybrid Model Based on Neural Network Quantile Regression and Kernel Density Estimation

Ex-situ deformation measurement system based on dual-reflector