首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于RGB颜色空间和随机矩形区域的显著性检测方法
引用本文:杨波,姚志均,王金武.一种基于RGB颜色空间和随机矩形区域的显著性检测方法[J].舰船电子对抗,2013(5):51-55.
作者姓名:杨波  姚志均  王金武
作者单位:船舶重工集团公司723所,扬州225001
摘    要:提出了一种基于RGB颜色空间和随机矩形区域的显著性检测方法.该方法以R、G、B作为图像特征,然后随机产生不同位置和大小的矩形区域,并统计每个矩形区域内各像素特征值与该区域的特征均值之间的距离,再综合所有矩形区域和所有特征得到最终的显著图.因不需进行颜色空间转换,可大幅减少计算时间;同时,RGB颜色空间三通道的亮度变化比较一致,使得在特征融合时能够充分利用所有特征的信息,因而取得了更好的检测效果.实验结果表明该方法能更快速、更有效地检测出图像中的显著性区域.

关 键 词:视觉注意  显著性检测  RGB颜色空间  随机矩形区域

A Saliency Detection Method Based on RGB Color Space and Random Rectangular Regions
YANG Bo,YAO Zhi-jun,WANG Jin-wu.A Saliency Detection Method Based on RGB Color Space and Random Rectangular Regions[J].Shipboard Electronic Countermeasure,2013(5):51-55.
Authors:YANG Bo  YAO Zhi-jun  WANG Jin-wu
Affiliation:(The 723 Institute of CSIC,Yangzhou 225001 ,China)
Abstract:This paper proposes a saliency detection method based on RGB color space and random rectangular regions. In this method, the R,G and B channels are the image feature. Then the rectan- gular regions of different position and size are randomly generated, and for each rectangular region, the distances between the pixel's feature value and the average feature value of the region are com- puted. Finally, all rectangular regions and features are combined to yield the final saliency map. Since the proposed method does not require the conversion of color spaces, the computing time is greatly reduced. At the same time,the brightness changes of three channels in the RGB color space are relatively consistent, which is helpful to make full use of the information of all features in the process of feature fusion and achieve better detection result. The experimental results show that the proposed method can detect the salient region of the image more quickly and effectively.
Keywords:visual attention  saliency detection  RGB color space  random rectangular region
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号