The method integrates a sensitivity analysis to characterize the impact of the variations in the parameters of a CAD model on the evolution of the deviation between the CAD model itself and the point cloud to be fitted. This paper addresses the way a SA-based Reverse Engineering technique can be enhanced by identifying its optimal default setting parameters for the fitting of CAD geometries to point clouds of digitized parts. However, parameter setting is a key factor for its performance, but it is also awkward work.
The experimental results show that the model achieves excellent performance in parts classification and prediction.ĭue to its capacity to evolve in a large solution space, the Simulated Annealing (SA) algorithm has shown very promising results for the Reverse Engineering of editable CAD geometries including parametric 2D sketches, 3D CAD parts and assemblies. It includes the calculation of weights and the construction of confusion matrix values.
We adopt this model to conduct a series of experiments on the Pacon dataset.
It can effectively solve the problems of high-cost and error-prone manual parts classification and greatly reduce the cost of automatic detection and classification of enterprises. The method can utilize a large amount of data of mechanical parts and then analyze and learn from the data through convolutional neural networks to detect and classify parts more accurately. Therefore, this paper proposes a part classification method based on a convolutional neural network. Different types and sizes of parts make automatic segmentation of parts on industrial lines prone to wrong segmentation. However, the current automated production lines for mechanical parts face great challenges. Mechanical parts are an important part of the machinery industry. Computational costs and required time are at the moment considerable. Accuracy of reconstructed models is comparable/better than state of the art results. Fitting process is controlled by a Particle Swarm Optimization algorithm. A feature-based parametric-associative modelling history is retrieved. Highlights A novel CAD reconstruction method fitting a CAD template model to mesh data.
Five reconstruction tests, covering both synthetic and real-scanned mesh data, are presented and discussed in the manuscript the results are finally compared with models generated by state of the art reverse engineering software and key aspects to be addressed in future work are hinted at. The proposed implementation exploits a cooperation between a CAD software package (Siemens NX) and a numerical software environment (MATLAB). As a result, a parametric CAD model that perfectly fulfils the imposed geometric relations is produced and a feature tree, defining an associative modelling history, is available to the reverse engineer. The CAD template is fitted upon the mesh data, optimizing its dimensional parameters and positioning/orientation by means of a particle swarm optimization algorithm.
The reconstruction process is performed relying on a CAD template, whose feature tree and geometric constraints are defined according to the a priori information on the physical object. CAD template) of the object to be reconstructed in order to retrieve a meaningful digital representation, a novel reverse engineering approach for the reconstruction of CAD models starting from 3D mesh data is proposed. Inspired by recent works suggesting the possibility/opportunity of exploiting a parametric description (i.e. Template-Based reverse engineering approaches represent a relatively poorly explored strategy in the field of CAD reconstruction from polygonal models.