全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于BP神经网络的图像局部模糊测量

DOI: 10.11834/jig.20150103

Keywords: BP神经网络,模糊测量,奇异值,DCT系数

Full-Text   Cite this paper   Add to My Lib

Abstract:

目的目前模糊测量方法难以处理存在纹理平坦区域时的局部模糊测量。针对该问题,提出一种基于BP(backpropagation)神经网络的图像局部模糊检测方法。方法该方法采用所有奇异值以描述不同尺度信息随模糊变化情况,并与描述高频信息变化的DCT(discretecosinetransform)非零系数结合,实现奇异值向量和DCT非零系数个数联合的混合模糊测度,达到空频联合的模糊度描述。在此基础上,通过训练BP神经网络分类器,实现图像块模糊值预测。结果单幅局部模糊图像实验中,较好区分纹理平坦区域和模糊区域的模糊程度;多幅局部模糊图像的统计实验中,召回率-准确率(RP)评估曲线显示在相同召回率下准确率较其他方法高。结论该方法可以较准确地实现局部模糊图像(特别是存在纹理平坦区域的局部模糊图像)的模糊测量。

References

[1]  Marziliano P, Dufaux F, Winkler S, et al. A no-reference perceptual blur metric[C]//Proceedings of the 9th International Conference on Image Processing. New York, USA: IEEE, 2002, 3:57-60.[DOI: 10.1109/ICIP.2002.1038902]
[2]  Tong H, Li M, Zhang H, et al. Blur detection for digital images using wavelet transform[C]//Proceedings of the 10th International Conference on Multimedia and Expo. Taipei, Taiwan, China: IEEE, 2004: 17-20. [DOI: 10.1109/ICME.2004.1394114]
[3]  Narvekar N D, Karam L J. A no-reference image blur metric based on the cumulative probability of blur detection [J].IEEE Transactions on Image Processing, 2011, 20(9): 2678-2683. [DOI: 10.1109/TIP.2011.2131660]
[4]  Soleimani S, Rooms F, Philips W. Efficient blur estimation using multi-scale quadrature filters [J]. Signal Processing, 2013, 93(7): 1988-2002. [DOI: 10.1016/j.sigpro.2012.12.015]
[5]  Grimaldi D, Kurylyak Y, Lamonaca F. Detection and parameters estimation of locally motion blurred objects[C]//Proceedings of the 6th International Conference on Intelligent Data Acquisition and Advanced Computing Systems. Prague, Czech: IEEE, 2011: 483-487. [DOI: 10.1109/IDAACS.2011.6072800]
[6]  Levin A. Blind motion deblurring using image statistics [C]//Proceedings of the Conference on Advances in Neural Information Processing Systems. Vancouver, Canada:MIT Press, 2006:841-848.[DOI:10.1.1.110.2995]
[7]  Chakrabarti A, Zickler T, Freeman W T. Analyzing spatially-varying blur[C]//Proceedings of the 23rd International Conference on Computer Vision and Pattern Recognition. San Francisco,USA:IEEE,2010: 2512-2519.[DOI: 10.1109/CVPR.2010.5539954]
[8]  Marichal X, Ma W Y, Zhang H J. Blur determination in the compressed domain using DCT information[C]//Proceedings of the 6th International Conference on Image Processing. Kobe, Japan:IEEE, 1999,2: 386-390. [DOI: 10.1109/ICIP.1999.822923]
[9]  Su B, Lu S, Tan C L. Blurred image region detection and classification[C]//Proceedings of the 19th ACM Iternational Conference on Multimedia. Toronto, Canada:ACM, 2011: 1397-1400.[DOI: 10.1145/2072298.2072024]
[10]  Hsu P, Chen B Y. Blurred image detection and classification [C]//Proceedings of 14th International Conference on Advances in Multimedia Modeling. Kyoto,Japan: ACM Press, 2008,4903: 277-286.[DOI: 10.1007/978-3-540-77409-9_26]
[11]  Liu R, Li Z, Jia J. Image partial blur detection and classification[C]//Proceedings of the 21st Computer Vision and Pattern Recognition. Anchorage,USA:IEEE,2008:1-8.[DOI: 10.1109/CVPR.2008.4587465]
[12]  Kanchev V, Tonchev K, Boumbarov O. Blurred image regions detection using wavelet based histograms and SVM[C]//Proceedings of the 6th International Conference on Intelligent Data Acquisition and Advanced Computing Systems. Prague, Czech: IEEE, 2011, 1: 457-461. [DOI: 10.1109/IDAACS.2011.6072795]
[13]  Andrews H, Patterson C. Singular value decompositions and digital image processing [J]. IEEE Transactions on Speech and Signal Processing, 1976, 24(1): 26-53.[DOI: 10.1109/TASSP.1976.116276]
[14]  Cong S. Neural Network Theory and Applications with MATLAB Toolboxes [M]. 3rd ed. Hefei: Press of University of Science and Technology of China, 2009:63-65.[丛爽.面向 MATLAB工具箱的神经网络理论与应用[M].3版.合肥:中国科学技术大学出版社, 2009:63-65.]
[15]  Yang J W. Digital image classification system based on BP neural network [D].Wuhan:Huazhong University of Science and Technology,2012.[杨建伟,基于BP神经网络的数字图像分类系统研究[D]. 武汉:华中科技大学,2012.]
[16]  Chang C C, Lin C J. LIBSVM: a library for support vector machines [J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2(3): #27. [DOI: 10.1145/1961189.1961199]

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133