吉林大学学报(工学版) ›› 2013, Vol. 43 ›› Issue (05): 1408-1414.doi: 10.7964/jdxbgxb201305041

• 论文 • 上一篇    下一篇

基于IPDSH兴趣点空间区域划分的图像检索

张旭, 郭宝龙, 孟繁杰, 孙伟   

  1. 西安电子科技大学 智能控制与图像工程研究所, 西安 710071
  • 收稿日期:2012-10-29 出版日期:2013-09-01 发布日期:2013-09-01
  • 通讯作者: 郭宝龙(1962- ),男,教授,博士生导师.研究方向:模式识别与智能系统,图像处理和图像通信.E-mail:blguo@xidian.edu.cn E-mail:blguo@xidian.edu.cn
  • 作者简介:张旭(1984- ),男,博士研究生.研究方向:图像检索.E-mail:xuzhang@mail.xidian.edu.cn
  • 基金资助:

    国家自然科学基金项目(61003196,61201290);中央高校基本科研业务费专项基金项目(K50510040008,K50512040011,K5051304024).

Image retrieval based on IPDSH and region division

ZHANG Xu, GUO Bao-long, MENG Fan-jie, SUN Wei   

  1. Institute of Intelligent Control & Image Engineering, Xidian University, Xi'an 710071, China
  • Received:2012-10-29 Online:2013-09-01 Published:2013-09-01

摘要:

针对基于兴趣点的传统图像检索方法的不足,提出了一种利用兴趣点检测和空间区域划分的图像检索新方法。首先使用一种结合SIFT和Harris特性的尺度空间兴趣点检测算法(IPDSH)来检测图像的稳定兴趣点;然后利用稳定兴趣点的空间位置对图像进行环形和凸包区域划分,并计算凸包内的颜色直方图和环形区域中稳定兴趣点邻域内伪泽尼克矩;最后以两种特征的加权特征向量对图像进行检索。该方法实现简单,检索速度快,能保证检索算法对图像旋转、平移的鲁棒性,且有效减少了图像中不稳定兴趣点对检索带来的干扰,图像检索的准确度有效提高了7.0%~15.1%。

关键词: 信息处理技术, 图像检索, 兴趣点, 区域划分, 特征提取

Abstract:

The accuracy of the current interest points detection algorithms are usually influenced by the unstable interest points in the non-interest regions. To overcome this disadvantage, an image retrieval algorithm combining interest point detection and region division is proposed. First, the interest points detection algorithm, combining the feature of Scale-Invariant Feature Transform (SIFT) and Harris (IPDSH), detects the stable interest points. Second, using the spatial positions of the stable interest points, the algorithm divides the image into annular region and convex hull, and calculates the color histogram of pixels in the convex hull and the pseudo-Zernike in the local field of stable interest points in the annular region. Finally, the algorithm retrieves the image with the weighted feature vector. Experimental results show that the proposed method possesses high retrieval speed, simple realization, robustness in rotation and translation, and can reduce the impact of unstable interest points. The image retrieval precision can by improve by 7.0%-15.1%.

Key words: information processing technology, image retrieval, interest points, region division, feature extraction

中图分类号: 

  • TN919.8

[1] Schmid C, Mohr R. Local gray value invariants for image retrieval[J]. IEEE Transaction on PAMI, 1997, 19(5):530-535.

[2] Mikoljczyk K, Schmid C. Indexing based on scale-Invariant feature[C]//Proc of International Conference on Computer Vision, Vancouver, 2001.

[3] Shao H, Ferrari V, Svoboda T, et al. Fast indexing for image retrieval based on local appearance with re-ranking[C]//Proceedings of IEEE International Conference on Image Processing, Barcelona,Spain,2003.

[4] 汪华章,何小海,宰文姣. 基于局部和全局特征融合的图像检索[J]. 光学精密工程,2008,16 (6): 1098-1104. Wang Hua-zhang, He Xiao-hai, Zai Wen-jiao. Image retrieval based on combining local and global features[J]. Optics and Precision Engineering, 2008, 16(6): 1098-1104.

[5] 符祥,曾接贤. 基于兴趣点匹配和空间分布的图像检索方法[J]. 中国激光,2010,37(3):774-778. Fu Xiang, Zeng Jie-xian. A novel image retrieval method based on interest points matching and distribution[J]. Chinese Journal of Lasers, 2010, 37(3):774-778.

[6] 陈绵书,杨树媛,赵志杰,等. 多点多样性密度算法及其在图像检索中的应用[J]. 吉林大学学报:工学版,2011,41(5): 1456-1460. Chen Mian-shu, Yang Shu-yuan, Zhao Zhi-jie,et al. Multi-points diverse density learning algorithm and its application in image retrieval[J]. Journal of Jilin University(Engineering and Technology Edition), 2011, 41(5): 1456-1460.

[7] Zheng X, Zhou M, Wang X C. Interest point based medical image retrieval[C]//Lecture Notes in Computer Science.Beijing:Springer Verlag, 2008.

[8] 陈慧婷,覃团发,唐振华,等. 综合纹理统计模型与全局主颜色的图像检索方法[J]. 北京邮电大学学报,2011, 34:100-103, 118. Chen Hui-ting, Qin Tuan-fa, Tang Zhen-hua, et al. A method of image retrievals based on texture probability statistics and global dominant color[J]. Journal of Beijing University of Posts and Telecommunications, 2011, 34:100-103, 118.

[9] 曾智勇,张学军,崔江涛,等. 基于显著兴趣点颜色及空间分布的图像检索新方法[J]. 光子学报,2006,35(2):308-311. Zeng Zhi-yong, Zhang Xue-jun, Cui Jinag-tao, et al. A novel image retrieval algorithm based on color and distribution of prominent interest points[J]. Acta Photonica Sinca, 2006,35(2): 308-311.

[10] 全燕鸣,黎淑梅. 大型工件测量系统中的快速图像拼接方法[J]. 华南理工大学学报:自然科学版,2011,39 (8): 60-65. Quan Yan-ming, Li Shu-mei. Fast image mosaic method for large-scale workpiece measurement system[J]. Journal of South China University of Technology (Natural Science), 2011, 39 (8): 60-65.

[11] Preparata F P, Shamos M I. Computational Geometry: an Introduction[M]. New York: Springer-Verlag, 1985.

[12] 孟繁杰,郭宝龙. 一种基于兴趣点颜色及空间分布的图像检索方法[J]. 西安电子科技大学学报:自然科学版,2005,32(2):256-259. Meng Fan-jie, Guo Bao-long. A novel image retrieval algorithm based on the color and distribution of interest points[J]. Journal of Xidian University(Natural Science), 2005, 32(2):256-259.

[13] 王向阳,陈景伟,于永健. 一种基于彩色边缘综合特征的图像检索方法[J]. 华南理工大学学报,2010,23(2):216-221. Wang Xiang-yang, Chen Jing-wei, Yu Yong-jian. Edge-based color image retrieval by using multiple features[J]. Pattern Recognition and Artificial Intelligence,2010, 23(2):216-221.

[14] Wu J H, Wei Z R, Li C Y. Color and texture feature for content based image retrieval[J]. International Journal of Digital Content Technology and Its Applications, 2010, 4(3):43-49.

[1] 苏寒松,代志涛,刘高华,张倩芳. 结合吸收Markov链和流行排序的显著性区域检测[J]. 吉林大学学报(工学版), 2018, 48(6): 1887-1894.
[2] 徐岩,孙美双. 基于卷积神经网络的水下图像增强方法[J]. 吉林大学学报(工学版), 2018, 48(6): 1895-1903.
[3] 黄勇,杨德运,乔赛,慕振国. 高分辨合成孔径雷达图像的耦合传统恒虚警目标检测[J]. 吉林大学学报(工学版), 2018, 48(6): 1904-1909.
[4] 李居朋,张祖成,李墨羽,缪德芳. 基于Kalman滤波的电容屏触控轨迹平滑算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1910-1916.
[5] 应欢,刘松华,唐博文,韩丽芳,周亮. 基于自适应释放策略的低开销确定性重放方法[J]. 吉林大学学报(工学版), 2018, 48(6): 1917-1924.
[6] 陆智俊,钟超,吴敬玉. 星载合成孔径雷达图像小特征的准确分割方法[J]. 吉林大学学报(工学版), 2018, 48(6): 1925-1930.
[7] 刘仲民,王阳,李战明,胡文瑾. 基于简单线性迭代聚类和快速最近邻区域合并的图像分割算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1931-1937.
[8] 单泽彪,刘小松,史红伟,王春阳,石要武. 动态压缩感知波达方向跟踪算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1938-1944.
[9] 姚海洋, 王海燕, 张之琛, 申晓红. 双Duffing振子逆向联合信号检测模型[J]. 吉林大学学报(工学版), 2018, 48(4): 1282-1290.
[10] 全薇, 郝晓明, 孙雅东, 柏葆华, 王禹亭. 基于实际眼结构的个性化投影式头盔物镜研制[J]. 吉林大学学报(工学版), 2018, 48(4): 1291-1297.
[11] 耿庆田, 于繁华, 王宇婷, 高琦坤. 基于特征融合的车型检测新算法[J]. 吉林大学学报(工学版), 2018, 48(3): 929-935.
[12] 陈绵书, 苏越, 桑爱军, 李培鹏. 基于空间矢量模型的图像分类方法[J]. 吉林大学学报(工学版), 2018, 48(3): 943-951.
[13] 陈涛, 崔岳寒, 郭立民. 适用于单快拍的多重信号分类改进算法[J]. 吉林大学学报(工学版), 2018, 48(3): 952-956.
[14] 孟广伟, 李荣佳, 王欣, 周立明, 顾帅. 压电双材料界面裂纹的强度因子分析[J]. 吉林大学学报(工学版), 2018, 48(2): 500-506.
[15] 林金花, 王延杰, 孙宏海. 改进的自适应特征细分方法及其对Catmull-Clark曲面的实时绘制[J]. 吉林大学学报(工学版), 2018, 48(2): 625-632.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!