基于随机漫步图匹配的三维CAD模型相似评价

Similarity assessment of 3D CAD model based on random walks graph matching

为了提高三维CAD模型检索的效率,提出了基于随机漫步图匹配的相似评价算法。使用图转化函数消除特征依赖图的末端节点;使用概率归一化函数消除特征间的不合理干扰,实现了对三维CAD模型的高效相似评价。实验结果表明,该方法具有较好的有效性和较高的效率。

Aiming at enhancing the efficiency of 3D CAD model retrieval,a new similarity assessment algorithm based on random-walks was proposed.Graph translation function was used to eliminate the sink nodes in feature dependency graph.Probability normalization function was used to eliminate the unreasonable disturbance among features.3D CAD models were efficiently retrieved.Experimental results demon-strate that the method is efficient and effective.