Abstract
The main objective of blind image de-blurring is to recover a sharp image from a given blurry image. A good estimation of the kernel plays an important role in recovering a sharp image. However, if the local object textures are neglected when the kernel is being estimated, this can lead to over-smoothing or can produce a strong ringing effect. In this paper, a new image regularization term based on the Probability Weighted Moments (PWM) for kernel estimation is proposed named as Probability Weighted Moments Regularization (PWMR). PWMR has the ability to preserve the small local texture structure in an image while minimizing the artifacts. Further, it can preserve the better contrast information between neighboring pixels and their corresponding central pixels in a current sliding window; moreover, it has the ability to resist outliers even in a small sample size. The kernel estimated by PWMR is subsequently used to recover the sharp latent image. An extensive comparison of synthetic and real standard benchmark images indicates the effectiveness of PWMR compared to current state-of-the-art blind image de-blurring methods.
Similar content being viewed by others
References
Beck, A.; Teboulle, M.: A fast iterative shrinkage- thresholding algorithm for linear inverse problems. SIAM J Imag Sci, 2, pp. 183–202(2009)
Chanand TF, Wong C-K (1998) Total variation blind deconvolution. IEEE Trans Image Process 7:370–375
Cho S.; and Lee, S.: Fast motion deblurring. In ACM Trans Graph (TOG), 28, p. 145(2009)
Dawood H, Dawood H, Guo P (2012) Combining the contrast information with WLD for texture classification. IEEE Int Conf Comput Sci Auto Eng (CSAE) 2012:203–207
Downton F (1966) Linear estimates with polynomial coefficients. Biometrika 53:129–141
Fergus R, Singh B, Hertzmann A, Roweis ST, Freeman WT (2006) Removing camera shake from a single photograph. ACM Trans Graphics (TOG) 25:787–794
Jiangxin D, Pan J, Su Z, Yang M (2017) Blind image deblurring with outlier handling. Proc IEEE Conf Comput Vision Pattern Recogn IEEE Conf Comput Vision Pattern Recogn (CVPR) 2017:2478–2486
Jinsha P, Deqing S, Hanspeter P, Hsuan YM (2016) Blind image deblurring using dark channel prior. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2016:1628–1636
Jinsha P, Deqing S, Hanspeter P, Hsuan YM (2017) Deblurring images via Dark Channel prior. IEEE Trans Pattern Anal Mach Intell (PAMI)
Krishnan D, Fergus R (2009) Fast image deconvolution using hyper-Laplacian priors. Adv Neural Inform Process Syst (NIPS) 2009:1033–1041
Krishnan D, Tay T, Fergus R (2011) Blind deconvolution using a normalized sparsity measure. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2011:233–240
Lai WS, Ding JJ, Lin YY, Chuang YY (2015) Blur kernel estimation using normalized color-line priors. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2015:64–72
Levin A, Weiss Y (2011) F. Durand, Freeman, W. T.: efficient marginal likelihood optimization in blind deconvolution. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2011:2657–2664
Levin A, Fergus R, Durand F, Freeman W (2007) Image and depth from a conventional camera with a coded aperture. ACM Trans Graph (TOG) 26:70
Levin A, Weiss L, Durand F, Freeman WT (2009) Understanding and evaluating blind deconvolution algorithms. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2009:1964–1971
Lian J, Zheng Y, Jiao W, Yan F, Zhao B (2018) Deblurring sequential ocular images from multi-spectral imaging (MSI) via mutual information. Med Biol Eng Comput 56(6):1107–1113
Michaeli T, Irani M (2014) Blind deblurring using internal patch recurrence. Eur Conf Comput Vision (ECCV) 2014:783–798
Mohammad T, Li Y, Monga V (2018) Blind image Deblurring using row-column sparse representations. IEEE Signal Process Lett (SPL) 25:273–278
Muhammad F, Riaz M (2006) Probability weighted moments approach to quality control charts. Econ Qual Contrl 21:251–260
Muhammad F, Aslam M, Pasha GR (2008) Adaptive estimation of heteroscedastic linear regression model using probability weighted moments. J Mod Appl Stat Methods 7:15
Perrone D, Favaro P (2014) Total variation blind deconvolution: the devil is in the details. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2014:2909–2916
Pu H, Fan M, Yang J, Lian J (2018) Quick response barcode deblurring via doubly convolutional neural network. Multimed Tools Appl, pp.1–16
Shan Q, Jia J, Agarwala A (2008) High-quality motion deblurring from a single image. ACM Trans Graph (TOG) 27:73
Singh D, Kumar V (2017) Modified gain intervention filter based dehazing technique. J Modern Optics (JMO) 64:2165–2178
Singh D, Kumar V (2017) Dehazing of remote sensing images using fourth-order partial differential equations based trilateral filter. IET Comput Vis
Singh D, Kumar V (2018) Defogging of road images using gain coefficient-based trilateral filter. J Electron Imag 27:013004
Whyte O, Sivic J, Zisserman A, Ponce J (2012) Non-uniform deblurring for shaken images. Int J Comput Vision (IJCV) 98:168–186
Wipf D, Zhang H (2013) Analysis of Bayesian blind deconvolution. Int Workshop Energy Minim Meth Comput Vision Pattern Recogn 2013:40–53
Wipf D, Zhang H (2014) Revisiting bayesian blind deconvolution. J Mach Learn Res: 3595–3634
Xu L, Jia L (2010) Two-phase kernel estimation for robust motion deblurring. In European Conference on Computer Vision (ECCV) 2010:157–170
Xu L, Zheng S, Jia J (2013) Unnatural l0 sparse representation for natural image deblurring. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2013:1107–1114
Yue T, Cho S, Wang J, Dai Q (2014) Hybrid image deblurring by fusing edge and power spectrum information. Eur Conf Comput Vision (ECCV) 2014:79–93
Zhang H, Wipf D, Zhang Y (2013) Multi-image blind deblurring using a coupled adaptive sparse prior. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2013:1051–1058
Zhong DL, Cho S, Metaxas D, Paris S, Wang J (2013) Handling noise in single image deblurring using directional filters. IEEE Conf Comput Vision Pattern Recogn (CVPR) 2013:612–619
Zhou Y, Komodakis N (2014) A map-estimation framework for blind deblurring using high-level edge priors. Eur Conf Comput Vision (ECCV) 2014:142–157
Zuo W-M, Dongwei R, David Z, Shuhang G, Lei Z (2016) Learning iteration-wise generalized shrinkage–thresholding operators for blind deconvolution. IEEE Trans Image Process (TIP) 25:1751–1764
Acknowledgements
This work is fully supported by the grants from the Joint Re-search Fund in Astronomy (Grant No. U1531242) under cooperative agreement between the National Natural Science Foundation of China (NSFC) and Chinese Academy of Sciences (CAS), Prof. Ping Guo is the author to whom all correspondence should be addressed.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Dawood, H., Dawood, H., Ping, G. et al. Probability weighted moments regularization based blind image De-blurring. Multimed Tools Appl 79, 4483–4498 (2020). https://doi.org/10.1007/s11042-019-7520-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-019-7520-9