具有目标偏见的全局对比度显著性区域检测

系统仿真学报 ›› 2015, Vol. 27 ›› Issue (10): 2489-2496.

具有目标偏见的全局对比度显著性区域检测

蔡强^1,2, 薛子育^1,2, 毛典辉^1,2, 李海生^1,2

1.北京工商大学,计算机与信息工程学院,北京 100048;
2.北京工商大学,食品安全大数据技术北京市重点实验室,北京 100048

收稿日期:2015-06-13 修回日期:2015-07-24 出版日期:2015-10-08 发布日期:2020-08-07
作者简介:蔡强(1969-),男,重庆,博士,教授,研究方向为计算机图形学、计算几何、科学可视化、智能信息处理。
基金资助:
食品安全大数据技术北京市重点实验室专项基金(19008001069); 北京市属高等学校青年英才计划资助项目(YETP1452)

Salient Region Detection based on Object-Biased Gaussian Refinement and Global Contrast

Cai Qiang^1,2, Xue Ziyu^1,2, Mao Dianhui^1,2, Li Haisheng^1,2

1. School of Computer and Information Engineering, Beijing Technology and Business University, Beijing 100048, China;
2. Beijing Key Laboratory of Big Data Technology for Food Safety, Beijing Technology and Business University, Beijing 100048, China

Received:2015-06-13 Revised:2015-07-24 Online:2015-10-08 Published:2020-08-07

摘要/Abstract

摘要： 针对全局对比度的显著性检测算法在图像边缘处的物体检测不完全的缺点,提出一种具有目标偏见的全局对比度显著性检测方法。在基于图的分割以后,利用全局对比度计算显著性值。根据显著性物体出现的位置调整高斯模型中心。利用分割区块相对显著性物体的位置与全局对比度确定显著性值。文中方法考虑了全局对比度和显著物体的空间位置,全局对比度算法图像中心确定方式有了改进。理论分析和实验结果表明,该方法可以很好的适用于各类图像的显著性检测,主观效果得到改善,客观指标得到提高。

关键词: 目标偏见, 全局对比度, 显著区域检测, 中心偏见, 高斯模型

Abstract: In view of the disadvantages such as the incomplete detection in image boundary which is caused by the global contrast significance detection algorithm, a visual saliency detection algorithm was proposed named salient region detection based on object-biased Gaussian refinement and global contrast. After segmentation based on graph, the global contrast was used to calculate the saliency value of each segmentation blocks. According to the location of the salient object, the center of the Gaussian model was adjusted. Both saliency detection value and Gaussian model combined effect of the final saliency values. The method paid attention to the global contrast and the spatial position of a salient object and improving the way to determine the image center. Theoretical analysis and experiments demonstrate that the method has a better saliency result which can detect the salient object region of all kinds of image effectively. The subjective quality is improved obviously, and the objective indicators are improved partly.

Key words: object-biased, global contrast, salient region detection, center-biased, Gaussian model

中图分类号:

TP391.4

蔡强, 薛子育, 毛典辉, 李海生. 具有目标偏见的全局对比度显著性区域检测[J]. 系统仿真学报, 2015, 27(10): 2489-2496.

Cai Qiang, Xue Ziyu, Mao Dianhui, Li Haisheng. Salient Region Detection based on Object-Biased Gaussian Refinement and Global Contrast[J]. Journal of System Simulation, 2015, 27(10): 2489-2496.

参考文献

[1] H Peng, B Li, R Ji, et al. Salient object detection via low-rank and structured sparse matrix decomposition[C]// AAAI, 2013. USA: AIAA, 2013.
[2] Y-F Ma, H-J Zhang.Contrast-based image attention analysis by using fuzzy growing[C]// ACM Multimedia. USA: ACM, 2003: 374-381.
[3] J Harel, C Koch, P Perona.Graph-based visual saliency[C]// Conference on Neural Information Processing Systems, USA, 2006: 545-552.
[4] N Li, J Ye, Y Ji, et al. Saliency detection on light fields[C]// Conference on Computer Vision and Pattern Recognition, USA, 2014: 2806-2813.
[5] R Margolin, A Tal, L Zelnik-Manor.What makes a patch distinct?[C]// Conference on Computer Vision and Pattern Recognition, USA, 2013: 1139-1146.
[6] Mingming Cheng, Niloy J Mitra, Xiaolei Huang, et al. Global Contrast based Salient Region Detection[C]// IEEE Conference on Computer Vision and Pattern Recognition. USA: IEEE, 2011: 409-416.
[7] L Itti, C Koch, E Niebur.A model of saliency-based visual attention for rapid scene analysis[J]. IEEE TPAMI (S0162-8828), 1998, 20(11): 1254-1259.
[8] T Liu, Z Yuan, J Wang, et al.Learning to detect a salient object[J]. IEEE TPAMI (1468-0394), 2010, 33(2): 410.
[9] S K Mannan, C Kennard, M Husain.The role of visual salience in directing eye movements in visual object agnosia[J]. Current Biology (S0960-9822), 2009, 19(6): 247-248, 409.
[10] J Reynolds, R Desimone.Interacting roles of visual salience in v4[J]. Neuron (S0896-6273), 2003, 37(5): 853-863.
[11] X Li, Y Li, C Shen, et al.Contextual hypergraph modeling for salient object detection[C]// Sydney, NSW: IEEE International Conference on Computer Vision, 2013: 3328-3335.
[12] C Rother, V Kolmogorov, A Blake.“Grabcut”- Interactive foreground extraction using iterated graph cuts. ACNTrans[J]. Graph (S1093-3263), 2004, 23(3): 309-314, 413.
[13] U Rutishauser, D Walther, C Koch, et al. Is bottom-up attention useful for object recognition?[C]// IEEE Conference on Computer Vision and Pattern Recognition. USA: IEEE, 2004: 37-44.
[14] A Borji, D N Sihite, L Itti.Salient object detection: A benchmark[C]// ECCV. Germany: Springer Berlin Heidelberg, 2012: 414-429.
[15] X Hou, L Zhang.Saliency detection: A spectral residual approach[C]// IEEE Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2007: 1-8.