[1]曲怀敬*,王恒斌,徐佳,等.基于DTCWT域统计特征融合的纹理图像检索[J].山东建筑大学学报,2020,(03):28-35.[doi:10.12077/sdjz.2020.03.005]
 QU Huaijing*,WANG Hengbin,XU Jia,et al.Texture image retrieval based on statistical feature fusion in DTCWT domain[J].Journal of Shandong jianzhu university,2020,(03):28-35.[doi:10.12077/sdjz.2020.03.005]
点击复制

基于DTCWT域统计特征融合的纹理图像检索()
分享到:

《山东建筑大学学报》[ISSN:1673-7644/CN:37-1449/TU]

卷:
期数:
2020年03期
页码:
28-35
栏目:
研究论文
出版日期:
2020-06-15

文章信息/Info

Title:
Texture image retrieval based on statistical feature fusion in DTCWT domain
文章编号:
1673-7644(2020)03-0028-08
作者:
曲怀敬*王恒斌徐佳王纪委魏亚南
(山东建筑大学 信息与电气工程学院,山东 济南 250101)
Author(s):
QU Huaijing* WANG Hengbin XU Jia WANG Jiwei WEI Yanan
( School of Information and Electrical Engineering, Shandong Jianzhu University, Jinan 250101, China )
关键词:
双树复小波变换纹理图像检索统计特征特征融合
Keywords:
dual-tree complex wavelet transform texture image retrieval statistical feature feature fusion
分类号:
TP391
DOI:
10.12077/sdjz.2020.03.005
文献标志码:
A
摘要:
在多尺度变换域,将各子带系数的统计特征进行互补融合可以有效地提高纹理图像检索的性能。文章利用双树复小波变换提出一种新的将低频子带系数的能量特征、高频子带幅值系数的Weibull分布参数特征以及相对相位系数的wrapped Cauchy分布参数特征相融合的纹理图像检索方法,采用VisTex纹理图像库进行检索。结果表明:采用多类系数统计特征的互补融合,以及最优的相似性测度加权组合,能够显著地提高纹理图像检索系统的平均检索率;与现有的7种纹理图像检索方法相比较,所获得的较高平均检索率为86.74%。
Abstract:
In the multi-scale transform domain, the performance of texture image retrieval can be effectively improved by the complementary fusion of the statistical features of each sub-band coefficients. This paper proposes a new texture image retrieval method which combines the energy feature of the low-frequency subband coefficients, the Weibull distribution parameter feature of the high-frequency subband amplitude coefficients and the wrapped Cauchy distribution parameter feature of the relative phase coefficients by using the double-tree complex wavelet transform. The experimental results show that the average retrieval rate of texture image retrieval system can be significantly improved by using the complementary fusion of multi-class coefficients statistical features and the optimal weighted combination of similarity measures. Compared with the existing 7 texture image retrieval methods, the proposed approach achieves a higher average retrieval rate with 86.74%.

参考文献/References:

[1]周文罡, 李厚强, 田奇. 图像检索技术研究进展[J]. 南京信息工程大学学报, 2017, 9(6):613-634. [2]刘丽,赵凌君,郭承玉,等. 图像纹理分类方法研究进展和展望[J]. 自动化学报,2018, 44(4):584-607. [3]苗聪聪. 基于隐马尔可夫树模型与旋转不变性的遥感图像纹理检索方法研究[D].北京:中国矿业大学,2015. [4]兰蓉, 郭思忱, 贾世英. 基于纹理与形状特征融合的刑侦图像检索算法[J]. 计算机工程与设计, 2018, 39(4): 1106-1110. [5]Do M N, Vetterli M. Wavelet-based texture retrieval using gener-alized Gaussian density and Kullback-Leiblerdistance[J]. IEEE Trans Image Processing, 2002, 11(2):146-158. [6]Kwitt R, Uhl A. Lightweight probabilistic texture retrieval[J]. IEEE Trans Image Processing, 2010, 19(1): 241-253. [7][JP3]Yang H Y, Liang L L, Zhang C, et al. Weibull statistical modeling for textured image retrieval using nonsubsampledcontourlettransform[J].Soft Computing, 2019,23(13):4749-4764.[JP] [8]杨娟,李永福,王荣贵,等. 基于双广义高斯模型和多尺度融合的纹理图像检索方法[J].电子与信息学报,2016,38(11):2856-2863. [9]Vo A, Oraintara S, Nguyen N. Vonn distribution of relative phase for statistical image modeling in complex wavelet domain[J]. Signal Processing, 2011, 91(1): 114-125. [10]李朝荣,付波,林劼.小波域Copula多维模型纹理检索[J].中国科学:信息科学,2014,44(12):1527-1541. [11]刘爽. 多特征融合图像检索方法及其应用研究[D]. 哈尔滨:哈尔滨理工大学,2016. [12]殷明,王治成,于立萍. 基于NSST和RI-LPQ的纹理图像检索[J].计算机工程,2014,40(10):258-263,269. [13]Ves E D, Acevedo D, Ruedin A, et al. A statistical model for magnitudes and angles of wavelet frame coefficients and its application to texture retrieval[J]. Pattern Recognition, 2014, 47(9): 2925-2939. [14]Qu H J, Wang H B. Texture image retrieval based on directional filter selection and statistical modeling[C].Proceedings of Tenth International Conference on Graphics and Image Processing,International Society for Optics and Photonics, 2019, 11069: 1106923. [15]曲怀敬. 基于PDTDFB域混合建模的纹理图像检索[J]. 计算机辅助设计与图形学学报,2012,24(10):1329-1336. [16]Vo A, Oraintara S. A study of relative phase in complex wavelet domain: Property, statistics and applications in texture image retrieval and segmentation[J]. Signal Processing: Image Communication, 2010, 25(1): 28-46.

备注/Memo

备注/Memo:
收稿日期:2020-03-07 基金项目:山东省自然科学基金项目(ZR2014FM016);山东省重大科技创新工程项目(2019JZZY010120)作者简介:曲怀敬(1965-),男,副教授,博士,主要从事模式识别及多尺度、多方向图像处理等方面的研究.E-mail:quhuaijing@sdjzu.edu.cn[*通讯作者]
更新日期/Last Update: 2020-06-17