Duhwan Mun

Determination of appropriate level of detail of a three-dimensional computer-aided design model from a permissible dissimilarity for fully automated simplification:

The level of detail required for a three-dimensional computer-aided design model varies according to its purpose. Thus, it is necessary to develop an automated method for controlling the level of detail of three-dimensional computer-aided design models to facilitate the use of three-dimensional computer-aided design models in different application areas. When multiple three-dimensional computer-aided design models need to be simplified in a batch mode without user intervention, it would be advantageous if the appropriate level of details for the models were automatically determined. This study proposes a new and automated method to determine the appropriate level of detail of each three-dimensional computer-aided design model during the simplification phase. To achieve this goal, the concept of an “appropriate level of detail for simplification” was defined from the viewpoint of shape dissimilarity. After that, an algorithm was developed to determine the appropriate level of detail of a model by adopting the feature-based simplification method and the shape distribution–based comparison method. Finally, a prototype system was invented to verify the proposed method through experiments, in which the appropriate level of details for test cases were determined by the system.

User-assisted integrated method for controlling level of detail of large-scale B-rep assembly models

ABSTRACT In recent times, there is a growing demand for technologies that control the level of detail (LOD) of large-scale three-dimensional (3D) computer-aided design (CAD) models. In particular, in industrial plants and in the construction industry, there is a significant need for simplification of large-scale models, because the number of models to be dealt with is tremendously large and their complexity is high. The degree of automation is, however, very low in these industries. As a result, many additional costs arise in the process of manually regenerating the 3D CAD models by the users. To resolve this problem, this study suggests a way to control effectively the LOD of large-scale B-rep assembly models. The proposed method consists of filtering, to remove those parts that ought to be removed, and progressive control, to remove sequentially the features acquired by volume decomposition. The user intervenes in the progressive control stage and adjusts the LOD for each component of the assembly. Based on the proposed method, a prototype system was implemented and experiments were performed for two test cases. The data size of the two test cases decreased, in average, to 65% in the first stage and 85% in the second stage. It was also found that the proposed component-level LOD control method is suitable and practical for a large-scale model.