|
|
论文:2014,Vol:32,Issue(6):872-876 |
|
|
引用本文： |
|
|
余瑞星, 朱丹, 高路, 郭冬梅. 基于谱残差显著区域检测的图像分辨率归一化方法[J]. 西北工业大学学报 |
|
|
Yu Ruixing, Zhu Dan, Gao Lu, Guo Dongmei. Normalizing Image Resolution in Saliency Detection Based On Spectral Residual[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 基于谱残差显著区域检测的图像分辨率归一化方法 |
|
| 余瑞星1, 朱丹1, 高路2, 郭冬梅2 |
|
1. 西北工业大学 航天学院, 陕西 西安 710072;
2. 上海无线电设备研究所,上海 200090 |
| 摘要: |
| 根据人类视觉系统的特点,提出一种自适应归一化图像分辨率的预处理方法。该方法通过计算图像灰度分布情况,根据图像所包含显著区域大小的不同,自适应的确定图像分辨率,打破了谱残差算法无论什么图像均将其归一化为固定分辨率大小(64×64)的局限;同时该算法还能够根据图像中所包含目标的不同大小,完整地提取出所有显著性目标,有效克服了谱残差算法倾向于小目标检测的缺点。实验结果表明:相对于谱残差算法,新算法简单,耗时少于0.1 s,检测准确率提高约15%,可以得到显著区域更为全面的检测效果。 |
| 关键词:
视觉注意
谱残差
自适应归一化
分辨率预处理
|
|
| Normalizing Image Resolution in Saliency Detection Based On Spectral Residual |
|
| Yu Ruixing1, Zhu Dan1, Gao Lu2, Guo Dongmei2 |
|
|
| Abstract: |
| According to the characteristics of human visual system,an adaptive normalizing method for preprocess-ing the image resolution is proposed. We analyze the size of the saliency region by calculating the gray distributionof the input image; then we define the image resolution adaptively. This method can help to do away with the limita-tion of normalizing all images into a fixed resolution (64×64) and to avoid the usually committed error of underesti-mating the total size of all effectively saliency regions when using the spectral residual method. The experimental re-sults on natural images and their analysis show preliminarily that this approach can: (1) improve the detection ac-curacy by consuming the time of less than 0. 1 second,thus improving hit rate by 15%; (2) estimate correctly thetotal size of all saliency regions. |
| Key words:
image resolution
calculations
discrete Fourier transforms
efficiency
experiments
invariance
ma-trix algebra
program processors
statistics
adaptive normalization
resolution-preprocessing
spectralresidual
visual attention
|
|
| 收稿日期: 2014-04-12
修回日期:
|
| DOI: |
| 基金项目: 国家自然科学基金(61101191);航空基金(20130153003);西北工业大学基础研究基金(JC20120216);上海航天科技创新基金(SAST201342);西北工业大学本科毕业设计重点扶持项目资助 |
| 通讯作者:
Email： |
| 作者简介: 余瑞星(1978-),女,西北工业大学副教授,主要从事图像处理、目标识别与成像制导技术研究。
|
|
| 相关功能 |
|
|
|
|
| 作者相关文章 |
|
| 余瑞星 在本刊中的所有文章 |
| 朱丹 在本刊中的所有文章 |
| 高路 在本刊中的所有文章 |
| 郭冬梅 在本刊中的所有文章 |
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] Itti L,Koch C. Computational Modeling of Visual Attention[J]. Nature Neuroscience,2001,2: 194-203
[2] Hou X,Zhang L. Saliency Detection: A Spectral Residual Approach[C] ∥IEEE Conference on Computer Vision and PatternRecognition,2007: 1-8
[3] 张巧荣, 顾国昌, 刘海波, 等. 利用多尺度频域分析的图像显著区域检测[J]. 哈尔滨工程大学学报, 2010, 31(3):361-365Zhang Qiaorong, Gu Guochang, Liu Haibo, et al. Salient Region Detection Using Multi-Scale Analysis in the Frequency Dodoma[J]. Journal of Harbin Engineering University,2010,31(3): 361-365 (in Chinese)
[4] 刘娟妮, 彭进业, 李大湘, 王平. 基于谱残差和多分辨率分析的显著目标检测[J]. 中国图象图形学报, 2011,16(2):244-249Liu Juanni,Peng Jinye,Li Daxiang,Wang Ping. Detecting Salient Objects Based on Spectral Residual and Multi-Resolution[J]. Journal of Image and Graphics,2011, 16(2): 244-249 (in Chinese)
[5] Ruderman D L. The Statistics of Natural Images[J]. Network: Computation in Neural Systems,1994, 5(4): 517-548
[6] Srivastava A,Lee A B,Simoncelli E P,et al. On Advances in Statistical Modeling of Natural Images[J]. Journal of Mathemati-cal Imaging and Vision,2003,18(1): 17-33
[7] Hou Xiaodi. Xiaodi's Homepage[EB/OL]. (2008-09-03)[2009-03-20].http∥www.Klab. caltech.edu/～xhou/
[8] 赵倩, 曹家麟, 胡越黎. 结合高斯多尺度变换和颜色复杂度计算的显著区域检测[J]. 仪器仪表学报,2012,33(2):405-412Zhao Qian,Cao Jialin,Hu Yueli. Salient Region Detection Based on Gaussian Multi-Scale Transform and Color ComplexityMeasure[J]. Chinese Journal of Scientific Instrument,2012,33(2):405-412 (in Chinese) |
|
|
|
|
|
|
|
