期刊文献+

多尺度阴影图像局部精准检索仿真 被引量:2

Multiscale Shadow Image Local Accurate Retrieval Simulation
下载PDF
导出
摘要 对多尺度阴影图像图像进行局部检索,可以满足用户搜索所有尺度上的模糊图像位置。对多尺度阴影图像局部的精准检索,需要引入非线性模糊成员函数,利用基于颜色信息熵的方法获取其权重,完成图像的局部检索。传统方法利用双树复数小波变换获得各子带系数,融合阴影模糊图像各个子带系数特征,但忽略了对模糊函数权重的求取,导致检测精度偏低。提出基于视觉注意机制的模糊图像局部检索方法。将局部空间引用模糊图像划分为多个不均匀的图像块,设定不同的权值,提取各个模糊图像子块的特征向量。计算阴影模糊图像颜色和边缘方向的色差直方图,将其作为描述阴影模糊图像的特征。在计算特征直方图时引入非线性模糊成员函数,利用基于颜色信息熵的方法动态分配特征直方图的权重。实验结果表明,该方法适用于连续性颜色过渡、梯度方向变化,模糊图像局部检索结果质量较高。 The aim of this article is to overcome defect of traditional local search method of multi-scale shadow image, such as poor detection precision. Based on mechanism of visual attention, a new local search method is pro- posed. Firstly, the blurred image was introduced, local space was divided into many uneven image blocks and differ- ent weights were set. Then, feature vector of each blurred image sub-block was extracted. Histogram of chromatic aberration of color and edge direction of shadow-blurred image was calculated and the histogram was used to describe feature of the blurred image. Nonlinear fuzzy member function was introduced during calculating feature histogram and weight of the feature histogram was allocated dynamically via method based on color information entropy. Simulation results show that the method is suitable for continuity color transition and change of gradient orientation. Local search result of the blurred image has high quality.
作者 马阿曼 吴薇 MA A-man;WU Wei(Wuyi University,College of Mathematics and Computer Science,Wuyishan Fujian 354300,Chin)
出处 《计算机仿真》 北大核心 2018年第7期335-338,413,共5页 Computer Simulation
基金 福建省科技厅引导性项目(2016N0030)
关键词 多尺度 阴影图像 局部检索 Multi scale Shadow image Local search
  • 相关文献

参考文献11

二级参考文献95

  • 1邱方鹏,冯玉才,梁俊杰.基于纹理和形状的图像相关反馈检索[J].计算机应用,2005,25(4):775-777. 被引量:8
  • 2周明全,韦娜,耿国华.交互信息理论及改进的颜色量化方法在图像检索中的应用研究[J].小型微型计算机系统,2006,27(7):1331-1334. 被引量:5
  • 3DAHANE G M, VISHWAKARMA S. Content based image retrieval system [ J]. International Journal of Engineering and Innoval:ive Technology, 2012, 1(5): 92-96.
  • 4LAI C-C, CHEN Y-C. A user-oriented image retrieval system based on interactive genetic algorithm [ J]. IEEE Transactions on Instru- mentation and Measurement, 2011, 60(10): 3318-3325.
  • 5WU Y, WU Y. Shape-based image retrieval using combining global and local shape features [ C]//CISP '09: Proceedings of the 2nd In- teruational Congress on Image and Signal Processing. Piscataway: IEEE, 2009:1-5.
  • 6RUI Y, HUANG T S, MEHROTRA S. Content-based image retriev- al with relevance feedback in MARS [ C] // Proceedings of the 1997 IEEE International Cmfference on Image Processing. Piscataway: IEEE, 1997,2:815-818.
  • 7DELP E J, MITCHELL 0 R. Image compression using block trun- cation coding [ J]. IEEE Transactions on Communications, 1979, 27(9) : 1335 - 1342.
  • 8YAP P T, PARAMESRAN R, OMG S H. Image analysis by Kraw- tchouck moments [ J]. IEEE Transactions on Image Processing, 2003, 12(11) : 1367 - 1377.
  • 9SUDHIR R, BABOO L D S S. An efficient CBIR technique with YUV color space and texture features [ J]. Computer Engineering and Intelligent Systems, 2011, 2(6): 778-785.
  • 10BOUNTHANH M, HAMAMOTO K, ATTACHOO B, et al. Con- tent-based image retrieval system based on combined and weighted multi-features [ C]//Proceedings of the 13th International Sympo- sium on Communications and Information Technologies. Piscat- away: IEEE, 2013:449-453.

共引文献52

同被引文献21

引证文献2

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部