Measuring instruments are intended to be intelligent, precise, multi-functional and developing multidirectionally, scientific, and reasonable; the reliable evaluation of measurement uncertainty of precision instruments is also becoming more and more difficult, and the evaluation of the Coordinate Measuring Machines (CMM) measurement uncertainty is among the typical problems. Based on Geometric Product Specification (GPS), this paper has systematically studied the CMM uncertainty for evaluating the size and geometrical errors oriented toward measurement tasks, and thus has realized the rapid and reliable evaluation of the CMM uncertainty for task-oriented measurement. For overestimation of the CMM uncertainty for task-oriented measurements in the initial evaluation, a systematic optimization solution has been proposed. Finally, the feasibility and validity of the evaluation model and the optimization method have been verified by three different types of measurement examples of diameter, flatness and perpendicularity. It is typical and representative to systematically solve the problem of the CMM uncertainty for evaluating the measurement tasks targeted at dimensions and geometric errors, and the research contents can be effectively applied to solve the uncertainty evaluation problems of other precision instruments, which are of great practical significance not only for promoting the combination of modern uncertainty theory and practical applications but also for improving the application values of precision measurement instruments.

Based on the Bayesian principle, the modern uncertainty evaluation methods can fully integrate prior and current sample information, determine the prior distribution according to historical information, and deduce the posterior distribution by integrating prior distribution and the current sample data with the Bayesian model. As such, it is possible to evaluate uncertainty, updating in real time the uncertainty of the measuring instrument according to regular measurement, and timely reflect the latest information on the accuracy of the measurement system. Based on the Bayesian information fusion and statistical inference principle, the model of uncertainty evaluation is established. The maximum entropy principle and the hill-climbing search optimization algorithm are introduced to determine the prior distribution probability density function and the sample information likelihood function. The probability density function of posterior distribution is obtained by the Bayesian formula to achieve the optimization estimation of uncertainty. Three methods of measurement uncertainty evaluation based on Bayesian analysis are introduced: the noninformative prior, the conjugate prior, and the maximum entropy prior distribution. The advantages and limitations of each method are discussed.

To overcome incomplete point cloud data obtained from laser scanners scanning complex surfaces, multi-viewpoint cloud data needs to be aligned for use. A hybrid improved SAC-IA with a KD-ICP algorithm is proposed for local point cloud alignment optimization. The scanned point cloud data is preprocessed with statistical filtering, as well as uniform down-sampling. The sampling consistency initial alignment (SAC-IA) algorithm is improved by introducing a dissimilarity vector for point cloud initial alignment. In addition, the iterative closest point (ICP) algorithm is improved by incorporating bidirectional KD-tree to form the KD-ICP algorithm for fine point cloud alignment. Finally, the algorithms are compared in terms of runtime and alignment accuracy. The implementation of the algorithms is based on the Visual Studio 2013 software configurating point cloud library environment for testing experiments and practical experiments. The overall alignment method can be 40%~50% faster in terms of running speed. The improved SAC-IA algorithm provides better transformed poses, combined with the KD-ICP algorithm to select the corresponding nearest neighbor pairs, which improves the accuracy, as well as the applicability of the alignment.

A structured light scanner is mainly used to collect three-dimensional (3D) data of an object; however, the uncertainty evaluation of a specific measurement task is not perfect. This study discusses the source of the uncertainty component of the measurement system of a structured blue-light scanner, including the instrument itself, data acquisition, data processing, and other factors. Using the surface profile as the evaluation object, PolyWorks software was used to perform the data processing of the point cloud denoising, splicing, and registration. An uncertainty evaluation is performed based on the black-box evaluation model. The experimental results show that the uncertainty components introduced by the point cloud stitching and point cloud registration are the main components of the entire measurement system. The bootstrap method does not need to consider the probability distribution type of the error sources, which improves computational efficiency. This method has several advantages when the distribution state is unknown. This work is conducive to improving the task-oriented uncertainty evaluation of 3D scanners.

The registration between the design model and the actual measurement model is a key problem in the detection of free-form surface parts. To improve the accuracy and robustness of free-form surface parts inspection, a Strategy Adaptive Particle Swarm Optimization (SAPSO) algorithm is proposed. Using the parameters composed of rotation and displacement variables as individual parameters, taking the minimum Euclidean distance between two models as the objective function and evolving from generation to generation, the calculation results can be guaranteed to be the global optimal solution. In SAPSO, an adaptive mechanism of inertia weight and velocity update strategy of particles is proposed to balance the global search and local exploration capabilities. The SAPSO algorithm is compared with other PSO variants. Simulation and actual measurement experiments show that the proposed algorithm is superior to other algorithms in terms of accuracy and robustness, and the effectiveness and practicability of SAPSO algorithm are verified.

Semantic image segmentation is an essential part of remote sensing image processing because accurate understanding of the ground information is the first step in obtaining useful knowledge of surface coverage. The popular semantic segmentation convolutional neural network model (DeepLab v3+) cannot effectively use attention information, resulting in coarse segmentation boundaries. In this work, a new type of bottleneck using attention information which can extract semantic information and more abundant features from images is proposed. Compared with original network, the model using new bottleneck finely segments the target regions, solves the problem of segmentation boundary roughness better, leading to higher mIoU and accuracy. Experimental results based on the dataset in the ISPRS benchmark on urban object classification show bringing attention model into semantic segmentation neural network improves performance.