基于哈希的二维工程CAD图纸检索技术
|
摘要
针对企业在设计创新过程中大量采用已有计算机辅助设计(CAD)图纸进行设计重用的情况,提出了一种基于哈希的二维工程CAD图纸检索方法。首先基于环形分割算法提取工程CAD图纸中每个组件对象几何特征;基于传统LBP算子提出了一种局部拓扑矢量量化模式(T-LVQP),实现对各个组件拓扑特征的提取;然后基于协方差描述符融合几何特征和拓扑特征,通过LBG算法将所有组件按照几何特征分组后得到工程CAD图纸的特征向量表达;最后通过迭代量化哈希算法生成图纸的哈希序列。实验结果表明,该算法检索速度快、准确度高,对于二维工程CAD图纸具有较好的检索效果。
In the process of design innovation, a large number of companies use existing computer aided design(CAD) drawings as the basis for design reuse. In this paper, hash algorithm is applied to the retrieval of 2 D engineering CAD drawings. The algorithm extracts geometry features of each component in the engineering CAD drawing based on ring partition, and the topology of local vector quantization pattern(T-LVQP) is proposed to extract the topological features of each component based on LBP operator. Then, all components in the drawings are classified into groups according to geometric features by the LBG algorithm, and the feature vector of each drawing is obtained via fused features based on the co-variance descriptor. Finally, the Hash sequence is generated by iterative quantization Hashing. The experimental results show that the algorithm has good performance in speed and accuracy for the retrieval of 2 D engineering CAD drawings.
In the process of design innovation, a large number of companies use existing computer aided design(CAD) drawings as the basis for design reuse. In this paper, hash algorithm is applied to the retrieval of 2 D engineering CAD drawings. The algorithm extracts geometry features of each component in the engineering CAD drawing based on ring partition, and the topology of local vector quantization pattern(T-LVQP) is proposed to extract the topological features of each component based on LBP operator. Then, all components in the drawings are classified into groups according to geometric features by the LBG algorithm, and the feature vector of each drawing is obtained via fused features based on the co-variance descriptor. Finally, the Hash sequence is generated by iterative quantization Hashing. The experimental results show that the algorithm has good performance in speed and accuracy for the retrieval of 2 D engineering CAD drawings.
引文
[1]黄晓剑.工程CAD中的快速绘制技术研究[D].北京:中国科学院计算技术研究所,2002.
[2]唐荣锡.CAD/CAM技术[M].北京:北京航空航天大学出版社,1994:1-4.
[3]PARK J H,UM B S.A new approach to similarity retrieval of 2-D graphic objects based on dominant shapes[J].Pattern Recognition Letters,1999,20(6):591-616.
[4]FONSECA M J,JORGE J A.Towards content-based retrieval of technical drawings through high-dimensional indexing[J].Computers&Graphics,2003,27(1):61-69.
[5]宋海玉,王鹏杰,王玲芬,等.基于内容的图纸检索的研究与设计[J].计算机工程与设计,2009,30(4):1017-1020.
[6]邱丽丽.基于内容的矢量工程图检索技术研究[D].南京:南京航空航天大学,2009.
[7]TANG Z,ZHANG X,LI X,et al.Robust image hashing with ring partition and invariant vector distance[J].IEEE Transactions on Information Forensics&Security,2015,11(1):200-214.
[8]周浪,苏智勇,唐卫清.面向二维工程CAD图的拓扑完整性校验算法[J].计算机辅助设计与图形学学报,2017,29(5):895-905.
[9]NANNI L,LUMINI A,BRAHNAM S.Survey on LBPbased texture descriptors for image classification[J].Expert Systems with Applications,2012,39(3):3634-3641.
[10]PAN Z,FAN H,ZHANG L.Texture classification using local pattern based on vector quantization[J].IEEETransactions on Image Processing,2015,24(12):5379-5388.
[11]LINDE Y,BUZO A,GRAY R M.An algorithm for vector quantizer design[J].IEEE Transactions on Communications,1980,28(1):84-95.
[12]TABIA H,LAGA H.Covariance-based descriptors for efficient 3D shape matching,retrieval,and classification[J].IEEE Transactions on Multimedia,2015,17(9):1591-1603.
[13]GONG Y,LAZEBNIK S.Iterative quantization:a procrustean approach to learning binary codes[C]//Computer Vision and Pattern Recognition.New York:IEEE Press,2011:817-824.
[14]张南.基于深度学习的图像哈希检索[D].天津:天津工业大学,2017.
[2]唐荣锡.CAD/CAM技术[M].北京:北京航空航天大学出版社,1994:1-4.
[3]PARK J H,UM B S.A new approach to similarity retrieval of 2-D graphic objects based on dominant shapes[J].Pattern Recognition Letters,1999,20(6):591-616.
[4]FONSECA M J,JORGE J A.Towards content-based retrieval of technical drawings through high-dimensional indexing[J].Computers&Graphics,2003,27(1):61-69.
[5]宋海玉,王鹏杰,王玲芬,等.基于内容的图纸检索的研究与设计[J].计算机工程与设计,2009,30(4):1017-1020.
[6]邱丽丽.基于内容的矢量工程图检索技术研究[D].南京:南京航空航天大学,2009.
[7]TANG Z,ZHANG X,LI X,et al.Robust image hashing with ring partition and invariant vector distance[J].IEEE Transactions on Information Forensics&Security,2015,11(1):200-214.
[8]周浪,苏智勇,唐卫清.面向二维工程CAD图的拓扑完整性校验算法[J].计算机辅助设计与图形学学报,2017,29(5):895-905.
[9]NANNI L,LUMINI A,BRAHNAM S.Survey on LBPbased texture descriptors for image classification[J].Expert Systems with Applications,2012,39(3):3634-3641.
[10]PAN Z,FAN H,ZHANG L.Texture classification using local pattern based on vector quantization[J].IEEETransactions on Image Processing,2015,24(12):5379-5388.
[11]LINDE Y,BUZO A,GRAY R M.An algorithm for vector quantizer design[J].IEEE Transactions on Communications,1980,28(1):84-95.
[12]TABIA H,LAGA H.Covariance-based descriptors for efficient 3D shape matching,retrieval,and classification[J].IEEE Transactions on Multimedia,2015,17(9):1591-1603.
[13]GONG Y,LAZEBNIK S.Iterative quantization:a procrustean approach to learning binary codes[C]//Computer Vision and Pattern Recognition.New York:IEEE Press,2011:817-824.
[14]张南.基于深度学习的图像哈希检索[D].天津:天津工业大学,2017.