计算机应用 ›› 2014, Vol. 34 ›› Issue (12): 3531-3535.

• 虚拟现实与数字媒体 • 上一篇    下一篇

基于滤波合成的关键显著性目标检测方法

王晨1,2,樊养余2,李波2,熊磊1   

  1. 1. 空军工程大学 航空航天工程学院, 西安 710038
    2. 西北工业大学 电子信息学院,西安 710072;
  • 收稿日期:2014-07-15 修回日期:2014-09-11 出版日期:2014-12-01 发布日期:2014-12-31
  • 通讯作者: 王晨
  • 作者简介:王晨(1977-),女,陕西高陵人,讲师,博士研究生,主要研究方向:图像显著性分析、图像分割、图像增强、特征检测;樊养余(1960-),男,陕西蓝田人,教授,博士,主要研究方向:数字图像处理、无线光通信、模式识别、虚拟现实;李波(1971-),男,陕西西安人,教授,博士,主要研究方向:宽带移动通信网络、多媒体无线互联网;熊磊(1975-),男,江西南昌人,副教授,博士,主要研究方向:数字图像处理、智能信息处理。
  • 基金资助:

    国家自然科学基金资助项目

Key salient object detection based on filtering integration method

WANG Chen1,2,FAN Yangyu2,LI Bo2,XIONG Lei1   

  1. 1. School of Aeronautics and Astronautics Engineering, Air Force Engineering University, Xi'an Shaanxi 710038, China
    2. School of Electronics and Information, Northwestern Polytechnical University, Xi'an Shaanxi 710072, China;
  • Received:2014-07-15 Revised:2014-09-11 Online:2014-12-01 Published:2014-12-31
  • Contact: WANG Chen

摘要:

针对显著性目标检测过程中的背景干扰问题,提出了一种基于滤波合成的关键显著性目标检测算法。该算法将局部指导滤波与改进的差分高斯(DoG)滤波方法相结合,使显著性目标更加凸显;然后,利用得到的显著性图确定关键点集合,通过调整因子得到更符合视觉机制的显著性检测结果。实验表明,所提算法优于现有显著性检测方法。与局部对比度(LC)方法、谱残差(SR)方法、基于直方图对比度(HC)方法、区域对比度(RC)方法、基于调频(FT)的方法等相比,背景与干扰目标得到有效抑制,同时具有更高的精度和更好的召回率。

Abstract:

Concerning the problem of the background interference during the salient object detection, a key salient object detection algorithm was proposed based on filtering integration in this paper. The proposed algorithm integrated the locally guided filtering with the improved DoG (Difference of Gaussia) filtering, and made the salient object more highlighted. Then, the key points set was determined by using the saliency map, and the result of saliency detection was got by adjustment factor, which was more suitable for human visual system. The experimental results show that the proposed algorithm is superior to existing significant detection methods. And it can restrain the background interference effectively, and have higher precision and better recall rate compared with other methods, such as Local Contrast (LC), Spectral Residual (SR), Histogram-based Contrast (HC), Region Contrast (RC) and Frequency-Tuned (FT).

中图分类号: