A progressive algorithm for block decomposition of solid models

At present, the dual structure-based block decomposition methods can generally obtain relatively ideal results. However, current dual structure-based block decomposition algorithms suffer from reliability and efficiency issues. To this end, to enable them to effectively deal with complex models, this paper proposes a progressive block decomposition algorithm. The algorithm first simplifies the input model by suppressing features and decomposes the simplified model into a block structure. Then, to recover the suppressed features in the simplified model's block structure, for each suppressed feature, the algorithm generates a local model to cover the feature. After decomposing the local model into a block structure with constraints, the algorithm replaces the corresponding block set in the block structure with the local model's block structure. In this step, the block structures are refined to obtain a consistency-ensured block structure. Finally, to achieve a balance between the total number of blocks and the blocks' quality in the block structure, the algorithm optimizes the consistency-ensured block structure by simplifying the structure. Experimental results show the effectiveness of the proposed method.

Automated summary

This paper proposes a progressive block decomposition algorithm that simplifies models by suppressing features and recovers the suppressed features to obtain a consistency-ensured block structure.