首页 > 论文 > 光学 精密工程 > 21卷 > 2期(pp:531-538)

视觉显著目标的自适应分割

Adaptive segmentation for visual salient object

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

基于视觉注意模型和最大熵分割算法, 提出了一种自适应显著目标分割方法来分离目标和复杂背景, 以便快速准确地从场景图像中检测出显著目标。首先, 通过颜色、强度、方向和局部能量4个特征通道获取图像的显著图; 通过引入局部能量通道来更好地描述了显著目标的轮廓。然后, 根据显著图中像素灰度的强弱构建不同的目标检测蒙板, 将每个蒙板作用于原图像作为预分割的结果, 再计算每个预分割图像的熵。最后, 利用最大熵准则估计图像目标熵,根据预分割图像的熵和目标熵判断选取最优显著目标分割图像。实验结果表明: 本文算法检测的显著目标更为完整, 分割性能F-measure达到0.56, 查全率和查准率分别为0.69和0.41, 相对于传统方法更为有效准确, 实现了在复杂背景下对显著目标的有效准确检测。

Abstract

On the basis of a visual attention model and a maximum entropy segmentation method, an adaptive segmentation method was proposed to segment the object from a complex background in the scene image and to detect a salient object effectively and accurately. First, the feature of original image was extracted via four channels on color, intensity, orientation and local energy. The profile of object feature was described more accurately by combining the channel of local energy with a simple biologically-inspired model. Then, object detection masks were constructed to remove background gradually according to the gray intensity of the pixels in the saliency map. By taking blend masks with the original image as a pre-segmentation result, the entropy of pre-segmentation images was computed. Finally, the entropy of salient object was estimated via maximization information entropy principle and the optimized image extraction for the salient object was obtained by estimating the relationship of entropy between salient object and masks in the saliency map. Experimental results indicate that the salient object detected by proposed method is more integrity, the F-measure of segmentation performance is 0.56, and the precision ratio and the recall ratio of detection are 0.69 and 0.41, respectively. The proposed method is more reasonable and effective than the traditional method, and it can satisfy the requirements of detecting the salient objects from complex backgrounds.

广告组6 - 调制器
补充资料

中图分类号:TP391

DOI:10.3788/ope.20132102.0531

所属栏目:信息科学

基金项目:国家自然科学基金资助项目(No.61101155); 吉林省科技发展计划资助项目(No.20101504)

收稿日期:2013-01-11

修改稿日期:2013-01-19

网络出版日期:--

作者单位    点击查看

赵宏伟:吉林大学 计算机科学与技术学院, 吉林 长春 130022
陈霄:吉林大学 计算机科学与技术学院, 吉林 长春 130022
刘萍萍:吉林大学 计算机科学与技术学院, 吉林 长春 130022
耿庆田:吉林大学 计算机科学与技术学院, 吉林 长春 130022

联系人作者:赵宏伟(zhaohw@jlu.edu.cn)

备注:赵宏伟 (1962-)男, 辽宁沈阳人,教授,博士生导师,主要从事智能信息系统与嵌入式技术方面的研究。

【1】丛明煜,何文家,逯力红, 等. 复杂背景成像条件下运动点目标的轨迹提取[J]. 光学 精密工程, 2012,20(7): 211-217.
CONG M Y,HE W J,LU LH, et al..Trace extraction of moving point targets in complex background images [J].Opt. Precision Eng.,2012,20(7): 211-217.(in Chinese)

【2】曾文静,万磊,张铁栋, 等.复杂海空背景下弱小目标的快速自动检测[J]. 光学 精密工程, 2012,20(2): 196-205.
ZHEN W J, WAN L, ZHANG T D, et al.. Fast detection of weak targets in complex sea-sky background [J].Opt. Precision Eng., 2012,20(2): 196-205.(in Chinese)

【3】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, 1998, 20(11):1254-1259.

【4】LI Z,LTTI L. Saliency and gist features for target detection in satellite images [J]. IEEE Transactions on Image Processing, 2011, 20(7): 2017-2029.

【5】HOU X D, HANG L Q. Saliency detection:a spectral residual approach [C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Minneapolis, USA: IEEE, 2007:1-8.

【6】CHENG M M,ZHANG G X,MITRA N J, et al.. Global contrast based salient region detection [C]. IEEE CVPR, Colorado Springs, USA,2011:409-416.

【7】胡正平, 孟鹏权. 全局孤立性和局部同质性图表示的随机游走显著目标检测算法 [J]. 自动化学报,2011,37(10):1279-1284.
HU ZH P, MENG P Q. Graph presentation random walk salient object detection algorithm based on global Isolation and local homogeneity [J].Acta Automatica Sinica,2011,37(10):1279-1284. (in Chinese)

【8】赵宏伟, 王慧, 刘萍萍, 等.有指向性的视觉注意计算机模型 [J].计算机研究与发展,2009, 46(7): 1192-1197.
ZHAO H W, WANG H, LIU P P, et al.. A computer model of directional visual attention [J].Journal of Computer Research and Development. 2009, 46(7):1192- 1197. (in Chinese)

【9】ANDREA P,MICHELA T,DOMENICO M, et al.. BOLD response to spatial phase congruency in human brain [J]. Vision of Journal, 2008, 8(10):1-15.

【10】LINDA H,AAPO H,SIMO V. Representation of cross-frequency spatial phase relationships in human visual cortex[J]. The Journal of Neuroscience, 2009, 29(45):14342-14351.

【11】VENKATESH S, OWENS R A. An energy feature detection scheme [C]. The International Conference on Image Processing, Singapore,1989: 553-557.

【12】MICHAEL F,GERALD S. The monogenic signal [J]. IEEE Transactions on Signal Proeessing,2001, 49: 3136-3144.

【13】许元男,赵远,刘丽萍,等.基于Renyi熵的显著图生成与目标探测[J]. 光学 精密工程,2010,18(3):723-731.
XUE Y N, ZHAO Y, LIU L P, et al.. Rényi entropy-based saliency map generation and target detection [J].Opt. Precision Eng., 2010, 18(3):723-731.(in Chinese)

【14】TAMAYO N,TRAVER V J. Entropy-based saliency computation in log-polar images [C].3rd International Conference on Computer Vision Theory and Applications, Funchal, Madeira, Portugal, 2008:501-506.

引用该论文

ZHAO Hong-wei,CHEN Xiao,LIU Ping-ping,GENG Qing-tian. Adaptive segmentation for visual salient object[J]. Optics and Precision Engineering, 2013, 21(2): 531-538

赵宏伟,陈霄,刘萍萍,耿庆田. 视觉显著目标的自适应分割[J]. 光学 精密工程, 2013, 21(2): 531-538

被引情况

【1】许志涛,刘金国,龙科慧,徐东,周怀得. 基于视觉注意机制的海洋监视卫星图像舰船目标检测. 激光与光电子学进展, 2013, 50(12): 121001--1

【2】王源源,黄大庆. 无人机侦察图像压缩. 光学 精密工程, 2014, 22(5): 1363-1370

【3】王刚,陈永光,杨锁昌,高敏,戴亚平. 采用图像块对比特性的红外弱小目标检测. 光学 精密工程, 2015, 23(5): 1424-1433

【4】王红玉,冯筠,崔磊,贺小伟,邱实. 应用显著纹理特征的医学图像配准. 光学 精密工程, 2015, 23(9): 2656-2665

【5】赵爱罡,王宏力,杨小冈,陆敬辉,姜 伟. 纹理粗糙度在红外图像显著性检测中的应用. 光学 精密工程, 2016, 24(1): 220-228

【6】成培瑞,王建立,王 斌,李正炜,吴元昊. 基于多尺度区域对比的显著目标识别. 中国光学, 2016, 9(1): 97-105

【7】乔 凯,陈 健,李中国,曾 磊,闫 镔. 锥束CT图像中的印刷电路板导线自动检测方法. 光学 精密工程, 2016, 24(2): 413-421

【8】杨会玲,柳红岩,王军,孙慧婷,何昕. 漂移扫描CCD星图的星点目标快速提取. 光电工程, 2016, 43(4): 15-19

【9】苏丽,吴俊杰,庞迪. 基于改进主动轮廓模型的全景海天线检测. 光学学报, 2016, 36(11): 1115003--1

【10】陈海永,徐 森,刘 坤,孙鹤旭. 基于谱残差视觉显著性的带钢表面缺陷检测. 光学 精密工程, 2016, 24(10): 2572-2580

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF