基于扩展轮廓的快速仿射不变特征提取

doi:10.13229/j.cnki.jdxbgxb20171157

摘要/Abstract

摘要：

针对快速仿射不变特征提取算法分类识别能力较差的问题，首先找出了算法证明及计算过程中的不足；其次提出并证明了扩展轮廓的仿射不变性，并通过扩展轮廓修正了算法中不足；然后，通过仿真实验给出了算法中参数设置的区间，并确定了算法中组合函数的个数及幂次。实验对比了基于扩展轮廓的快速仿射不变特征提取算法与原始算法及多尺度自卷积矩算法的分类识别能力，结果表明，基于扩展轮廓的快速仿射不变特征提取算法的分类识别能力明显优于原始算法和多尺度自卷积矩算法。

关键词: 计算机应用, 模式识别, 仿射不变特征, 扩展轮廓, 多尺度自卷积矩

Abstract:

To overcome the poor classification ability of the fast affine invariant feature extracting algorithm, an extended contour-based fast affine invariant feature extracting algorithm is developed. Firstly, the flaw in the proving and calculating procedures of the fast affine invariant feature extracting algorithm is proposed. Secondly, the definition of extended contour is proposed which is used to amend the fast affine invariant feature extraction algorithm. Thirdly, through simulation experiments, the parameter setting intervals in the algorithm are fixed and the numbers and powers of the component functions are determined. The simulation experiments are carried out to compared the proposed algorithm with the original fast affine invariant feature extracting algorithm and the multi-scale auto-convolution algorithm. The experimental results show that the proposed algorithm has better classification and recognition ability than the original fast affine invariant feature extracting algorithm and the multi-scale auto-convolution algorithm.

Key words: computer application, pattern recognition, affine invariant feature, extended contour, multi?scale auto convolution(MSA)

中图分类号:

TP391.4

翟凤文,党建武,王阳萍,金静,罗维薇. 基于扩展轮廓的快速仿射不变特征提取[J]. 吉林大学学报(工学版), 2019, 49(4): 1345-1356.

Feng⁃wen ZHAI,Jian⁃wu DANG,Yang⁃ping WANG,Jing JIN,Wei⁃wei LUO. Extended contour⁃based fast affine invariant feature extracting[J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(4): 1345-1356.

图/表 17

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

图15

表1

图16

参考文献 15

1	杨建伟, 李沛遥 . 基于分数阶矩的仿射不变特征提取[J]. 自动化学报, 2015, 41(12):2147⁃2154.
	Yang Jian-wei , Li Pei-yao . Affine invariant feature extraction based on fractional order moment[J]. Acta Electronica Sinica, 2015, 41(12):2147-2154
2	缪君, 储珺, 张桂梅 . 一种仿射不变的直线描述子与直线匹配[J]. 电子学报, 2015, 43(12):2505⁃2512.
	Miao Jun , Chu Jun , Zhang Gui⁃mei . An affine invariant line descriptor and line matching[J]. Chinese Journal of Electronics, 2015, 43(12): 2505⁃2512.
3	Heikkilä J . Multi⁃scale autoconvolution for affine invariant pattern recognition[C]∥The 16th International Conference on Pattern Recognition, Quebec City, Canada, 2002:10119.
4	Rahtu E , Salo M , Heikkila J . Affine invariant pattern recognition using multiscale autoconvolution[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2005, 27(6):908⁃918.
5	Kannala J , Rahtu E , Heikkila J . Affine registration with multi⁃scale autoconvolution[C]∥IEEE International Conference on Image Processing, Genova, Italy, 2005:1064⁃7.
6	Rahtu E , Salo M , Heikkilä J . Multiscale autoconvolution histograms for affine invariant pattern recognition[C]∥British Machine Vision Conference, Edinburgh, Uk, 2006:1059⁃1068.
7	Rahtu E , Salo M , Heikkil J , et al . Generalized affine moment invariants for object recognition[C]∥International Conference on Pattern Recognition, Hong Kong, 2006:634⁃637.
8	Hu Ming Kuei . Visual pattern recognition by moment invariants[J]. IRE Transactions on Information Theory, 1962, 8(2):179⁃187.
9	翟凤文, Muhammad Asim Azim, 王阳萍, 等 . Zernike矩快速算法的修正[J]. 吉林大学学报:工学版, 2014, 44(6):1860⁃1866.
	Zhai Feng⁃wen , Muhammad Asim Azim, Wang Yang⁃ping , et al . Amendment of Zernike moment fast computation[J]. Journal of Jilin University(Engineering and Technology Edition), 44(6): 1860⁃1866.
10	李晖晖, 滑立, 杨宁, 等 . 基于MSA特征和模拟退火优化的遥感图像多目标关联算法[J]. 吉林大学学报:工学版, 2015, 45(4):1353⁃1359.
	Li Hui⁃hui , Hua Li , Yang Ning , et al . Multi⁃target association algorithm for remote sensing images based on MSA features and simulated annealing optimization[J]. Journal of Jilin University(Engineering and Technology Edition), 2015, 45(4):1353⁃1359.
11	吴喆, 曾接贤, 高琪琪 . 显著图和多特征结合的遥感图像飞机目标识别[J]. 中国图象图形学报, 2017, 22(4):532⁃541.
	Wu Zhe ， Zeng Jie⁃xian ， Gao Qi⁃qi . Aircraft target recognition in remote sensing images based on saliency images and multi⁃feature combination[J]. Journal of Image and Graphics, 2017, 22(4):532⁃541.
12	张洁玉, 陈强, 白小晶, 等 . 仿射不变的多尺度自卷积熵提取方法[J]. 计算机辅助设计与图形学学报, 2009, 21(9):1328⁃1332.
	Zhang Jie⁃yu , Chen Qiang , Bai Xiao⁃jing ,et al . Affine invariant feature extraction based on multi⁃scale auto⁃convolution entropy[J]. Journal of Computer⁃Aided Design & Computer Graphics, 21(9):1328⁃1332.
13	Rahtu E , Salo M , Heikkilä J . A new efficient method for producing global affine invariants[J]. LNCS, 2005, 3617:407⁃414.
14	贾建华, 焦李成, 黄文涛 . 一种基于质心不变特性的仿射不变纹理特征提取算法[J]. 电子学报, 2008, 36(10):1910⁃1915.
	Jia Jian⁃hua , Jiao Li⁃cheng , Huang Wen⁃tao . Affine invariant texture feature extraction based on invariant centroid[J]. Acta Electronica Sinica, 2008, 36(10):1910⁃1915.
15	Petrou M , Kadyrov A . Features invariant to affine distortions from the trace transform[C]∥Proceeding of the IEEE Conference on Image Processing, Thessaloniki, Greece, 2001: 852⁃855.

相关文章 15

[1]	李雄飞,宋璐,张小利. 基于协同经验小波变换的遥感图像融合[J]. 吉林大学学报(工学版), 2019, 49(4): 1307-1319.
[2]	孙延君,申铉京,陈海鹏,赵永哲. 基于局部平面线性点的翻拍图像鉴别算法[J]. 吉林大学学报(工学版), 2019, 49(4): 1320-1328.
[3]	刘元宁,刘帅,朱晓冬,霍光,丁通,张阔,姜雪,郭书君,张齐贤. 基于决策粒子群优化与稳定纹理的虹膜二次识别[J]. 吉林大学学报(工学版), 2019, 49(4): 1329-1338.
[4]	李宾,申国君,孙庚,郑婷婷. 改进的鸡群优化算法[J]. 吉林大学学报(工学版), 2019, 49(4): 1339-1344.
[5]	王楠,李金宝,刘勇,张玉杰,钟颖莉. TPR⁃TF:基于张量分解的时间敏感兴趣点推荐模型[J]. 吉林大学学报(工学版), 2019, 49(3): 920-933.
[6]	刘富,宗宇轩,康冰,张益萌,林彩霞,赵宏伟. 基于优化纹理特征的手背静脉识别系统[J]. 吉林大学学报(工学版), 2018, 48(6): 1844-1850.
[7]	王利民,刘洋,孙铭会,李美慧. 基于Markov blanket的无约束型K阶贝叶斯集成分类模型[J]. 吉林大学学报(工学版), 2018, 48(6): 1851-1858.
[8]	金顺福,王宝帅,郝闪闪,贾晓光,霍占强. 基于备用虚拟机同步休眠的云数据中心节能策略及性能[J]. 吉林大学学报(工学版), 2018, 48(6): 1859-1866.
[9]	赵东,孙明玉,朱金龙,于繁华,刘光洁,陈慧灵. 结合粒子群和单纯形的改进飞蛾优化算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1867-1872.
[10]	刘恩泽,吴文福. 基于机器视觉的农作物表面多特征决策融合病变判断算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1873-1878.
[11]	欧阳丹彤, 范琪. 子句级别语境感知的开放信息抽取方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1563-1570.
[12]	刘富, 兰旭腾, 侯涛, 康冰, 刘云, 林彩霞. 基于优化k-mer频率的宏基因组聚类方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1593-1599.
[13]	桂春, 黄旺星. 基于改进的标签传播算法的网络聚类方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1600-1605.
[14]	刘元宁, 刘帅, 朱晓冬, 陈一浩, 郑少阁, 沈椿壮. 基于高斯拉普拉斯算子与自适应优化伽柏滤波的虹膜识别[J]. 吉林大学学报(工学版), 2018, 48(5): 1606-1613.
[15]	车翔玖, 王利, 郭晓新. 基于多尺度特征融合的边界检测算法[J]. 吉林大学学报(工学版), 2018, 48(5): 1621-1628.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

		J₁₂	J₁₃	J₁₄	J₂₃	J₂₄	J₃₄
T1	O	1.1032	2.7945	1.9064	0.4488	1.9761	8.0326
	N	5.5070	7.2457	2.1370	2.8211	5.7799	7.0127
	M	5.4017	8.3021	2.9541	2.2322	6.7721	10.5159
T2	O	5.0131	0.6587	0.9866	1.6523	9.3518	1.2623
	N	5.3153	0.7623	1.3552	1.7161	8.2898	1.5486
	M	4.6857	0.7613	1.6459	1.2520	5.2240	1.5440
T3	O	0.5680	1.2638	0.8324	0.8459	0.3213	0.6597
	N	0.6110	2.2677	1.8270	0.9857	0.3839	0.4931
	M	0.8256	2.1455	2.1840	0.2496	0.2407	0.5085
T4	O	0.8703	0.7001	0.4781	1.3752	0.877	0.2941
	N	1.1560	3.5281	1.6704	1.7401	0.4597	1.1943
	M	1.266	3.5481	1.8885	1.4718	0.2786	1.4369
T5	O	0.3155	0.3709	0.2800	0.4098	0.2011	0.3410
	N	0.6723	0.6052	0.1345	1.4938	0.7477	0.7438
	M	0.9683	0.7693	0.2938	1.6291	0.6796	0.9896
T6	O	1.2256	0.9845	0.5956	1.4401	0.2489	0.9891
	N	1.8734	1.7978	1.1511	3.3365	1.4667	3.0824
	M	2.1548	1.5891	1.3045	3.3482	1.6935	2.9658
T7	O	1.0424	1.7442	1.1964	1.5861	0.4114	1.6633
	N	0.6220	1.2677	0.4870	2.8803	0.7553	2.9218
	M	0.8511	0.8024	0.9422	2.5131	0.3957	3.4842
T8	O	9.4425	3.3529	1.9051	1.8982	1.5830	0.8370
	N	3.1828	11.2549	4.0737	7.2851	2.2636	2.8422
	M	6.7318	19.4526	4.5319	10.4487	1.5937	3.5344
T9	O	6.2368	0.5544	0.2710	0.9613	4.8863	0.4478
	N	4.7109	1.2118	0.6017	0.4666	1.9455	0.6242
	M	4.0699	1.1646	0.5917	0.4083	1.8814	0.6345
T10	O	1.3416	5.8264	5.6180	44.8413	9.1621	16.6656
	N	0.5326	4.6143	2.8479	50.9191	5.3271	10.3754
	M	0.3492	2.7033	1.9654	36.5806	7.6850	14.0528