基于多尺度轮廓段的形状特征提取与识别
|
摘要
形状识别是计算机视觉与模式识别领域的重要研究内容。形状的特征选取与描述是形状识别的研究热点。针对现有识别方法的不足,提出一种通过对不同长度轮廓段进行描述,进行特征提取的方法。对每个形状均在6种尺度下进行特征提取,每种尺度选取5种轮廓段特征参数,实现了对形状的特征描述。在形状识别阶段,使用动态时间规整(DTW)算法度量形状描述子之间的匹配距离,实现形状识别。分别在Kimia99、Kimia216和MPEG-7数据库中进行算法验证,结果表明基于多尺度轮廓段的形状特征描述子具有旋转、缩放、平移和局部遮挡不变性,识别率优于现有算法。
Shape recognition is an important research content in the field of computer vision and pattern recognition.Shape feature selection and description is a hot topic in shape recognition. In view of the shortcomings of the existing recognition methods, a method of feature extraction is proposed by describing different length contour segments, in which the feature parameters from six contour segments of the shape are achieved with five parameters for each contour segment.In the shape recognition stage, the Dynamic Time Warping(DTW)algorithm is used to measure the matching distance between shape descriptors to perform shape recognition. The method is verified in the Kimia99, Kimia216 and MPEG-7 databases. The results show that the shape feature descriptors based on multi-scale contour segments have invariance of rotation, scaling, translation and local occlusion, and the recognition prate is bettDyer than ?? theh existê ing m?ethods?. K
Shape recognition is an important research content in the field of computer vision and pattern recognition.Shape feature selection and description is a hot topic in shape recognition. In view of the shortcomings of the existing recognition methods, a method of feature extraction is proposed by describing different length contour segments, in which the feature parameters from six contour segments of the shape are achieved with five parameters for each contour segment.In the shape recognition stage, the Dynamic Time Warping(DTW)algorithm is used to measure the matching distance between shape descriptors to perform shape recognition. The method is verified in the Kimia99, Kimia216 and MPEG-7 databases. The results show that the shape feature descriptors based on multi-scale contour segments have invariance of rotation, scaling, translation and local occlusion, and the recognition prate is bettDyer than ?? theh existê ing m?ethods?. K
引文
[1]周瑜,刘俊涛,白翔.形状匹配方法研究与展望[J].自动化学报,2012,38(6):889-910.
[2] Shen W,Du C,Jiang Y,et al.Bag of shape features witha learned pooling function for shape recognition[J].Pattern Recognition Letters,2018,106:33-40.
[3]张桂梅,孙晓旭,章毅.基于小波变换的局部形状匹配[J].计算机工程与应用,2017,53(2):188-194.
[4]章毓晋.图像工程[M].3版.北京:清华大学出版社,2012.
[5] Zhang D,Lu G.Review of shape representation and descrip-tion techniques[J].Pattern Recognition,2004,37(1):1-19.
[6] Belongie S,Malik J,Puzicha J.Shape matching and objectrecognition using shape contexts[J].IEEE Transactions onPattern Analysis&Machine Intelligence,2006,2(4):509-522.
[7] Ling H,Jacobs D W.Shape classification using the inner-distance[J].IEEE Transactions on Pattern Analysis&MachineIntelligence,2007,29(2):286-299.
[8] Biswas S,Aggarwal G,Chellappa R.Efficient indexing forarticulation invariant shape matching and retrieval[C]//IEEE Conference on Computer Vision and Pattern Rec-ognition,2007:1-8.
[9] Yang C,Wei H,Yu Q,et al.A novel method for 2D non-rigid partial shape matching[J].Neurocomputing,2017,275.
[10] Alajlan N,Kamel M S,Freeman G.Multi-object imageretrieval based on shape and topology[J].Signal ProcessingImage Communication,2006,21(10):904-918.
[11]杨亚飞,郑丹晨,韩敏.一种基于多尺度轮廓点空间关系特征的形状匹配方法[J].自动化学报,2015,41(8):1405-1411.
[12] Attalla E,Siy P.Robust shape similarity retrieval basedon contour segmentation polygonal multiresolution andelastic matching[J].Pattern Recognition,2005,38(12):2229-2241.
[13]徐浩然,杨剑宇,黄伟国.等.形状的不变量特征提取与识别[J].中国图象图形学报,2017,22(8):1068-1078.
[14]王生生,郭湑,张家晨,等.基于全局与局部形状特征融合的形状识别算法[J].吉林大学学报(工学版),2016,46(5):1627-1632.
[15] Xu H,Yang J,Yuan J.Invariant multi-scale shape de-scriptor for object matching and recognition[C]//IEEEInternational Conference on Image Processing,2016:644-648.
[16] Shu X,Wu X J.A novel contour descriptor for 2D shapematching and its application to image retrieval[J].Image&Vision Computing,2011,29(4):286-294.
[17] Garc?a-Ordás M T,Alegre E,González-Castro V,et al.aZIBO:a new descriptor based in shape moments androtational invariant features[C]//International Conferenceon Pattern Recognition,2014:2395-2400.
[18]王江辉,吴小俊.基于形状轮廓特征的金字塔匹配算法[J].计算机工程与应用,2019,55(1):191-195.
[19]吴绍根,聂为清,路利军,等.形状的圆内距离变换[J].中国图象图形学报,2018,23(1):39-51.
[2] Shen W,Du C,Jiang Y,et al.Bag of shape features witha learned pooling function for shape recognition[J].Pattern Recognition Letters,2018,106:33-40.
[3]张桂梅,孙晓旭,章毅.基于小波变换的局部形状匹配[J].计算机工程与应用,2017,53(2):188-194.
[4]章毓晋.图像工程[M].3版.北京:清华大学出版社,2012.
[5] Zhang D,Lu G.Review of shape representation and descrip-tion techniques[J].Pattern Recognition,2004,37(1):1-19.
[6] Belongie S,Malik J,Puzicha J.Shape matching and objectrecognition using shape contexts[J].IEEE Transactions onPattern Analysis&Machine Intelligence,2006,2(4):509-522.
[7] Ling H,Jacobs D W.Shape classification using the inner-distance[J].IEEE Transactions on Pattern Analysis&MachineIntelligence,2007,29(2):286-299.
[8] Biswas S,Aggarwal G,Chellappa R.Efficient indexing forarticulation invariant shape matching and retrieval[C]//IEEE Conference on Computer Vision and Pattern Rec-ognition,2007:1-8.
[9] Yang C,Wei H,Yu Q,et al.A novel method for 2D non-rigid partial shape matching[J].Neurocomputing,2017,275.
[10] Alajlan N,Kamel M S,Freeman G.Multi-object imageretrieval based on shape and topology[J].Signal ProcessingImage Communication,2006,21(10):904-918.
[11]杨亚飞,郑丹晨,韩敏.一种基于多尺度轮廓点空间关系特征的形状匹配方法[J].自动化学报,2015,41(8):1405-1411.
[12] Attalla E,Siy P.Robust shape similarity retrieval basedon contour segmentation polygonal multiresolution andelastic matching[J].Pattern Recognition,2005,38(12):2229-2241.
[13]徐浩然,杨剑宇,黄伟国.等.形状的不变量特征提取与识别[J].中国图象图形学报,2017,22(8):1068-1078.
[14]王生生,郭湑,张家晨,等.基于全局与局部形状特征融合的形状识别算法[J].吉林大学学报(工学版),2016,46(5):1627-1632.
[15] Xu H,Yang J,Yuan J.Invariant multi-scale shape de-scriptor for object matching and recognition[C]//IEEEInternational Conference on Image Processing,2016:644-648.
[16] Shu X,Wu X J.A novel contour descriptor for 2D shapematching and its application to image retrieval[J].Image&Vision Computing,2011,29(4):286-294.
[17] Garc?a-Ordás M T,Alegre E,González-Castro V,et al.aZIBO:a new descriptor based in shape moments androtational invariant features[C]//International Conferenceon Pattern Recognition,2014:2395-2400.
[18]王江辉,吴小俊.基于形状轮廓特征的金字塔匹配算法[J].计算机工程与应用,2019,55(1):191-195.
[19]吴绍根,聂为清,路利军,等.形状的圆内距离变换[J].中国图象图形学报,2018,23(1):39-51.