联合灰色关联度和先验的图像显著性分析

系统仿真学报 ›› 2015, Vol. 27 ›› Issue (7): 1511-1519.

联合灰色关联度和先验的图像显著性分析

周强强^1,2, 王志成¹, 赵卫东¹, 陈宇飞¹, 王刚¹

1.同济大学电子与信息工程学院CAD研究中心,上海 201804;
2.江西农业大学计算机与信息工程学院,南昌 330027

收稿日期:2014-06-19 修回日期:2014-10-05 出版日期:2015-07-08 发布日期:2020-07-31
作者简介:周强强(1979-),男,江西永新,博士生,研究方向为计算机视觉;王志成(1975-),男,博士,副研究员,研究方向为模式识别。
基金资助:
国家自然科学基金项目资助(61103070)

Image Saliency Analysis Based on Grey Relational Computation and Prior Combination

Zhou Qiangqiang^1,2, Wang Zhicheng¹, Zhao Weidong¹, Chen Yufei¹, Wang Gang¹

1. CAD Research Center of College of Electronics & Information, TongJi University, Shanghai 201804, China;
2. College of Computer & Information, JiangXi Agriculture Universtiy, Nanchang 330027, China

Received:2014-06-19 Revised:2014-10-05 Online:2015-07-08 Published:2020-07-31

摘要/Abstract

摘要： 在基于边界背景先验,全局显著性先验和高层先验这三种信息基础上,提出了一种联合灰色关联度计算和先验知识的图像显著性分析方法。用SLIC方法对图像进行超像素分割,再分别选取图像4个边界的超像素构建参考序列,对图像剩余区域构建比较序列。进行灰色关联度计算,得到4个基于边界背景先验的初始显著性图,进行加权融合优化。进一步综合全局的颜色空间分布信息和高层先验信息,生成最终显著图。在公开的MSRA-1000数据集上进行试验,并与现有的几个经典方法进行了比较,显示有更好的准确率和召回率,且在较复杂背景下也有好的显著性检测效果。

关键词: 图像显著性, 灰色关联分析, 区域特征描述子, 先验知识

Abstract: An image saliency analysis method based on the combination of grey relational computation and prior knowledge, which is a saliency computation and integration based on three different priori of boundary background, global color distribution and top salient information. The SLIC method segment the image into superpixels, then selecting the boundary superpixels on four image sides respectively to construct the reference sequence, and the same method used to construct the comparative sequence of the rest superpixels. Then four initial saliency images was derived based on boundary background priori through a grey relational computation, to make weighted and improved. Furtherly integrated the global color spatial distribution information and the top priori information to produce the final saliency image. An experiment were made on the open MSRA-1000 datasets and compared the results with several existing classic methods, conclusions shows the method has a better precision and recall, even under a more complex background.

Key words: image saliency, grey relational analysis, regional characteristics descriptor, prior knowledge

中图分类号:

TP391

周强强, 王志成, 赵卫东, 陈宇飞, 王刚. 联合灰色关联度和先验的图像显著性分析[J]. 系统仿真学报, 2015, 27(7): 1511-1519.

Zhou Qiangqiang, Wang Zhicheng, Zhao Weidong, Chen Yufei, Wang Gang. Image Saliency Analysis Based on Grey Relational Computation and Prior Combination[J]. Journal of System Simulation, 2015, 27(7): 1511-1519.

参考文献

[1] Itti L, Koch C, Niebur E.A model of saliency-based visual attention for rapid scene analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence(S0162-8828), 1998, 20(11): 1254-1259.
[2] Zhang L, Tong M, Marks T, ,et al. SUN: a Bayesian framework for saliency using natural statistics [J]. Journal of Vision. SUN: a Bayesian framework for saliency using natural statistics [J]. Journal of Vision (S1534-7362), 2008, 8(7): 32: 1-20.
[3] Gao D, Han S, Vasconcelos N.Discriminant saliency, the detection of suspicious coincidences, and applications to visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2009, 31(6): 989-1005.
[4] Borji A, Itti L.State-of-the-art in visual attention modeling[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence(S0162-8828), 2013, 35(1): 185-207.
[5] Shen X, Wu Y.A unified approach to salient object detection via low rank matrix recovery[C]// 2012 IEEE Conference on. USA: IEEE, 2012: 853-860.
[6] Yang C, Zhang L, Lu H, et al. Saliency detection via graph-based manifold ranking[C]// Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. USA: IEEE, 2013: 3166-3173.
[7] Jiang Z, Davis L S.Submodular Salient Region Detection[C]// Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. USA: IEEE, 2013: 2043-2050.
[8] Liu R, Cao J, Lin Z, et al. Adaptive partial differential equation learning for visual saliency detection[C]// Computer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on. IEEE, 2014: 3866-3873.
[9] Wei Y, Wen F, Zhu W, et al. Geodesic saliency using background priors[M]// Computer Vision-ECCV 2012. Germany: Springer Berlin Heidelberg, 2012: 29-42.
[10] Jiang H, Wang J, Yuan Z, et al. Salient object detection: A discriminative regional feature integration approach[C]// Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. USA: IEEE, 2013: 2083-2090.
[11] Achanta R, Shaji A, Smith K, et al. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence(S0162-8828), 2012, 34(11): 2274-2282.
[12] 马苗, 张艳宁, 赵健. 灰色理论及其在图像工程中的应用[M]. 清华大学出版社, 2011.
[13] Shen X, Wu Y.A unified approach to salient object detection via low rank matrix recovery[C]// Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. USA: IEEE, 2012: 853-860.
[14] Liu T, Yuan Z, Sun J, et al. Learning to detect a salient object[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2011, 33(2): 353-367.
[15] Hou X, Zhang L.Saliency detection: A spectral residual approach[C]// Computer Vision and Pattern Recognition, 2007. CVPR'07, IEEE Conference on. USA: IEEE, 2007: 1-8.
[16] Harel J, Koch C, Perona P.Graph-based visual saliency[C]// Advances in Neural Information Processing Systems. 2006: 545-552.
[17] Achanta R, Hemami S, Estrada F, et al. Frequency-tuned salient region detection[C]// Computer Vision and Pattern Recognition, 2009. CVPR 2009, IEEE Conference on. USA: IEEE, 2009: 1597-1604.
[18] Boykov Y, Kolmogorov V.An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2004, 26(9): 1124-1137.