The Visual Computer

, Volume 33, Issue 2, pp 151–161 | Cite as

Automatic estimation and segmentation of partial blur in natural images

  • Taiebeh Askari JavaranEmail author
  • Hamid Hassanpour
  • Vahid Abolghasemi
Original Article


Digital images may contain undesired blurred regions. Automatic detection of such regions and estimation of the amount of blurriness in a given image are important issues in many computer vision applications. This paper presents a simple and effective method to automatically detect blurred regions. The proposed method consists of two main parts. First, a novel blur metric, which can significantly distinguish blur and non-blur regions, is proposed. This metric is then used to generate a blur map to encode the amount of blurriness for individual pixels in a given image. Finally, the estimated blur map is used to segment the input image into blurred/non-blurred regions by applying a pixon-based technique. The proposed approach is evaluated for out-of-focus and motion-blurred natural images. By conducting experiments on a large dataset containing real images with defocus blur and partial motion blur regions, qualitative and quantitative measures are performed. The obtained results in this paper show that the proposed approach outperforms the state-of-the-art methods for blur estimation in digital images.


Partial blur Blur metric Blur map Blur/non-blur region segmentation Pixon 


  1. 1.
    Tai, Y., Du, H., Brown, M.S., Lin, S.: Correction of spatially varying image and video motion blur using a hybrid camera. IEEE Trans. Pattern Anal. 32(6), 1012–1028 (2010)CrossRefGoogle Scholar
  2. 2.
    Xu, L., Jia, J.: Two-phase kernel estimation for robust motion deblurring. In: Proceedings of the 11th European Conference on Computer Vision (ECCV), pp. 157–170 (2010)Google Scholar
  3. 3.
    Levin, A., Fergus, R., Durand, F.: Image and depth from a conventional camera with a coded aperture. Assoc. Comput. Mach. Trans. Graph. 26(3), Article 70 (2007)Google Scholar
  4. 4.
    Shan, Q., Jia, J., Agarwala, A.: High-quality motion deblurring from a single image. Assoc. Comput. Mach. Trans. Graph. 27(3), 1–10 (2008)Google Scholar
  5. 5.
    Cho, S., Lee, S.: Fast motion deblurring. Assoc. Comput. Mach. Trans. Graph. 28(5), 1–8 (2009)Google Scholar
  6. 6.
    Fergus, R., Singh, B., Hertzmann, A., et al.: Removing camera shake from a single photograph. Assoc. Comput. Mach. Trans. Graph. 25(3), 787–794 (2006)Google Scholar
  7. 7.
    Krishnan, D., Tay, T., Fergus, R.: Blind deconvolution using a normalized sparsity measure. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 233–240 (2011)Google Scholar
  8. 8.
    Rugna, J.D., Konik, H.: Automatic blur detection for meta-data extraction in content-based retrieval context. Proc. SPIE 5304, 285–294 (2003)CrossRefGoogle Scholar
  9. 9.
    Levin, A.: Blind motion deblurring using image statistics. In: Proceedings of the 20th Annual Conference on Neural Information Processing Systems (NIPS ’06), pp. 841–848 (2006)Google Scholar
  10. 10.
    Dai, Sh., Wu, Y.: Estimating space-variant motion blur without deblurring. In: Proceedings of the 15th IEEE International Conference on Image Processing, pp. 661–664 (2008)Google Scholar
  11. 11.
    Chakrabarti, A., Zickler, T., Freeman, W.T.: Analyzing spatially-varying blur. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2512–2519 (2010)Google Scholar
  12. 12.
    Wang, W., Zheng, J., Zhou, H.: Segmenting, removing and ranking partial blur. Signal Image Video Process. 8(4), 647–655 (2014)CrossRefGoogle Scholar
  13. 13.
    Graf, F., Kriegel, H.P., Weiler, M.: Robust segmentation of relevant regions in low depth of field images. In: Proceedings of the 18th IEEE International Conference on Image Processing (ICIP), pp. 2861–2864 (2011)Google Scholar
  14. 14.
    Tai, Y.W., Brown, M.S.: Single image defocus map estimation using local contrast prior. In: Proceedings of the 16th IEEE International Conference on Image Processing (ICIP), pp. 1797–1800 (2009)Google Scholar
  15. 15.
    Kovacs, L., Szirnyi, T.: Focus area extraction by blind deconvolution for defining regions of interest. IEEE Trans. Pattern Anal. Mach. Intell. 29(6), 1080–1085 (2007)CrossRefGoogle Scholar
  16. 16.
    Kim, C.: Segmenting a low-depth-of-field image using morphological filters and region merging. IEEE Trans. Image Process. 14(10), 1503–1511 (2005)CrossRefGoogle Scholar
  17. 17.
    Li, H., Ngan, K.N.: Un-supervised video segmentation with low depth of field. IEEE Trans. Circuits Syst. Video Technol. 17(12), 1742–1751 (2007)CrossRefGoogle Scholar
  18. 18.
    Bahrami, K., Kot, A.C., Fan, I.: A novel approach for partial blur detection and segmentation. In: Proceedings of the IEEE International Conference on Multimedia and Expo (ICME), pp. 1–6 (2013)Google Scholar
  19. 19.
    Mavridaki, E., Mezaris, V.: No-reference blur assessment in natural images using Fourier transform and spatial pyramids. In: Proceedings of the IEEE International Conference on Image Processing (ICIP), pp. 566–570 (2014)Google Scholar
  20. 20.
    Liu, R., Li, Z., Jia, J.: Image partial blur detection and classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–8 (2008)Google Scholar
  21. 21.
    Chen, X., Yang, J., Wu, Q., et al.: Motion blur detection based on lowest directional high- frequency energy. In: Proceedings of the 17th IEEE International Conference on Image Processing, pp. 2533–2536 (2010)Google Scholar
  22. 22.
    Su, B., Lu, S., Tan, C.L.: Blurred image region detection and classification. In: Association for Computing Machinery (ACM) International Conference on Multimedia, pp. 1397–1400 (2011)Google Scholar
  23. 23.
    Liu, S., Wang, H., et al.: Automatic blur-kernel-size estimation for motion deblurring. Vis. Comput. Int. J. Comput. Graph. 31(5), 733–746 (2015)Google Scholar
  24. 24.
    Zhu, X., Cohen, S., Schiller, S., et al.: Estimating spatially varying defocus blur from a single image. IEEE Trans. Image Process. 22(12), 4879–4891 (2013)MathSciNetCrossRefGoogle Scholar
  25. 25.
    Zhuo, S., Sim, T.: Defocus map estimation from a single image. Int. J. Pattern Recognit. 44(9), 1852–1858 (2011)CrossRefGoogle Scholar
  26. 26.
    Shi, J., Xu, L., Jia, J.: Discriminative blur detection features. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2965–2972 (2014)Google Scholar
  27. 27.
    Chetouani, A., Mostafaoui, G., Beghdadi, A.: A new free reference image quality index based on perceptual blur estimation. In: Proceedings of the IEEE Pacific Rim Conference on Multimedia (PCM), pp. 1185–1196 (2009)Google Scholar
  28. 28.
    Crete, F., Dolmiere, T., Ladret, P., et al.: The blur effect: perception and estimation with a new no-reference perceptual blur metric. In: Proceedings of the SPIE, vol. 6492, no. 1, Article 0I, pp. 1–11 (2007)Google Scholar
  29. 29.
    Yang, F., Jiang, T.: Pixon-based image segmentation with Markov random fields. IEEE Trans. Image Process. 12(12), 1552–1559 (2003)CrossRefGoogle Scholar
  30. 30.
    Fauzi, M.F.A., Lewis, P.H.: A fully unsupervised texture segmentation algorithm. In: British Machine Vision Conference, pp. 519–528 (2003)Google Scholar
  31. 31.
    Discriminative Blur Detection Features (2014) Blur Detection Dataset. Accessed 29 April 2015

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Taiebeh Askari Javaran
    • 1
    Email author
  • Hamid Hassanpour
    • 1
  • Vahid Abolghasemi
    • 2
  1. 1.Image Processing and Data Mining (IPDM) Research Lab, Faculty of Computer Engineering and Information TechnologyUniversity of ShahroodShahroodIran
  2. 2.Faculty of Electrical Engineering and RoboticsUniversity of ShahroodShahroodIran

Personalised recommendations