Deblurring approach for motion camera combining FFT with α-confidence goal optimization

• Autorzy: Huang Lve , Wu Lushen , Xiao Wenyan , Peng Qingjin

Identyfikatory

DOI

10.37190/oa200202

• Języki publikacji: EN

Abstrakty

EN

Sharp images ensure success in the object detection and recognition from state-of-art deep learning methods. When there is a fast relative motion between the camera and the object being imaged during exposure, it will necessarily result in blurred images. To deblur the images acquired under the camera motion for high-quality images, a deblurring approach with relatively simple calculation is proposed. An accurate estimation method of point spread function is firstly developed by performing the Fourier transform twice. Artifacts caused by image direct deconvolution are then reduced by predicting the image boundary region, and the deconvolution model is optimized by an α-confidence statistics algorithm based on the greyscale consistency of the image adjacent columns. The proposed deblurring approach is finally carried out on both the synthetic-blurred images and the real-scene images. The experiment results demonstrate that the proposed image deblurring approach outperforms the existing methods for the images that are seriously blurred in direction motion.

Słowa kluczowe

EN

image deblurring; fast motion camera; confidence goal optimization; fast Fourier transform; high-railway defect detection

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Optica Applicata
• Rocznik: 2020
• Tom: Vol. 50, nr 2
• Strony: 185--198
• Opis fizyczny: Bibliogr. 30 poz., rys., tab.

Twórcy

autor

Huang Lve

• School of Mechanical and Electrical Engineering, Nanchang University, Nanchang, 330031, China
• College of Applied Science, Jiangxi University of Science and Technology, Ganzhou, 341000, China

autor

Wu Lushen

• School of Mechanical and Electrical Engineering, Nanchang University, Nanchang, 330031, China

autor

Xiao Wenyan

• Jiangxi University of Science and Technology, Nanchang, 330013, China

autor

Peng Qingjin

• Department of Mechanical Engineering, University of Manitoba, R3T2N2, Canada

Bibliografia

• [1] CHEONG J.Y., SIMON C., KIM C.-S., PARK I.K., Reflection removal under fast forward camera motion, IEEE Transactions on Image Processing 26(12), 2017, pp. 6061–6073, DOI: 10.1109/TIP.2017.2748389.
• [2] RUIZ P., ZHOU X., MATEOS J., MOLINA R., KATSAGGELOS A.K., Variational Bayesian blind image deconvolution: a review, Digital Signal Processing 47, 2015, pp. 116–127, DOI: 10.1016/j.dsp.2015.04.012.
• [3] GONG D., TAN M., SHI Q., VAN DEN HENGEL A., ZHANG Y., MPTV: matching pursuit-based total variation minimization for image deconvolution, IEEE Transactions on Image Processing 28(4), 2019, pp. 1851–1865, DOI: 10.1109/TIP.2018.2875352.
• [4] LU S., LIU Z., SHEN Y., Automatic fault detection of multiple targets in railway maintenance based on time-scale normalization, IEEE Transactions on Instrumentation and Measurement 67(4), 2018, pp. 849–865, DOI: 10.1109/TIM.2018.2790498.
• [5] YUAN X., WU L., PENG Q., An improved Otsu method using the weighted object variance for defect detection, Applied Surface Science 349, 2015, pp. 472–484, DOI: 10.1016/j.apsusc.2015.05.033.
• [6] HUANG L.E., WU L.S., CHENG H.W., Image blur types and parameters estimation using DCNN fusion with the LSTM, Journal of Basic Science and Engineering 26(5), 2018, pp. 1092–1100.
• [7] ZHANG J., PAN J., LAI W.-S., LAU R.W.M., YANG M.-H., Learning fully convolutional networks for iterative non-blind deconvolution, IEEE Conference on Computer Vision and Pattern Recognition, July 2017, Hawaii, USA, pp. 3817–3825, DOI: 10.1109/CVPR.2017.737.
• [8] DESHPANDE A.M., PATNAIK S., A novel modified cepstral based technique for blind estimation of motion blur, Optik 125(2), 2014, pp. 606–615, DOI: 10.1016/j.ijleo.2013.05.189.
• [9] CHANG M.M., TEKALP A.M., ERDEM A.T., Blur identification using the bispectrum, IEEE Transactions on Signal Processing 39(10), 1991, pp. 2323–2325, DOI: 10.1109/78.91207.
• [10] LUCY L.B., An iterative technique for the rectification of observed distributions, The Astronomical Journal 79(6), 1974, pp. 745–754.
• [11] BASELICE F., FERRAIOLI G., AMBROSANIO M., PASCAZIO V., SCHIRINZI G., Enhanced Wiener filter for ultrasound image restoration, Computer Methods and Programs in Biomedicine 153, 2018, pp. 71–81, DOI: 10.1016/j.cmpb.2017.10.006.
• [12] BENTAHAR Y., AFIFI M., DALIMI H., AMAR S., Restoration image degraded by a blurred variable in the field, Optica Applicata 48(1), 2018, pp. 5–14, DOI: 10.5277/oa180101.
• [13] DEMIRCAN-TUREYEN E., KAMASAK M.E., Directional total variation based image deconvolution with unknown boundaries, [In] Computer Analysis of Images and Patterns. CAIP 2017, [Eds.] Felsberg M., Heyden A., Krüger N., Lecture Notes in Computer Science, Vol. 10425, Springer, Cham, 2017, pp.473–484, DOI: 10.1007/978-3-319-64698-5_40.
• [14] ALMEIDA M.S.C., FIGUEIREDO M., Deconvolving images with unknown boundaries using the alternating direction method of multipliers, IEEE Transactions on Image Processing 22(8), 2013, pp. 3074–3086, DOI: 10.1109/TIP.2013.2258354.
• [15] BAI Y., CHEUNG G., LIU X., GAO W., Graph-based blind image deblurring from a single photograph, IEEE Transactions on Image Processing 28(3), 2019, pp. 1404–1418, DOI: 10.1109/TIP.2018.2874290.
• [16] KRISHNAN D., TAY T., FERGUS R., Blind deconvolution using a normalized sparsity measure, IEEE Conference on Computer Vision and Pattern Recognition, June 2011, Colorado, USA, pp. 233–240, DOI: 10.1109/CVPR.2011.5995521.
• [17] LEVIN A., WEISS Y., DURAND F., FREEMAN W.T., Efficient marginal likelihood optimization in blind deconvolution, IEEE Conference on Computer Vision and Pattern Recognition, June 2011, Colorado, USA, pp. 2657–2664, DOI: 10.1109/CVPR.2011.5995308.
• [18] XU L., ZHENG S., JIA J., Unnatural L0 sparse representation for natural image deblurring, IEEE Conference on Computer Vision and Pattern Recognition, June 2013, Oregon, USA, pp. 1107–1114, DOI: 10.1109/CVPR.2013.147.
• [19] SUN J., CAO W., XU Z., PONCE J., Learning a convolutional neural network for non-uniform motion blur removal, IEEE Conference on Computer Vision and Pattern Recognition, June 2015, Boston, Massachusetts, pp. 769–777, DOI: 10.1109/CVPR.2015.7298677.
• [20] HAN Y., KAN J., Blind color-image deblurring based on color image gradients, Signal Processing 155, 2019, pp. 14–24, DOI: 10.1016/j.sigpro.2018.09.032.
• [21] SHAMIK TIWARI, SHUKLA V.P., SINGH A.K., BIRADAR S.R., Review of motion blur estimation techniques, Journal of Image and Graphics 1(4), 2013, pp. 176–184, DOI: 10.12720/joig.1.4.176-184.
• [22] WANG Z., YAO Z., WANG Q., Improved scheme of estimating motion blur parameters for image restoration, Digital Signal Processing 65, 2017, pp. 11–18, DOI: 10.1016/j.dsp.2017.02.010.
• [23] YUAN L., SUN J., QUAN L., SHUM H.-Y., Image deblurring with blurred/noisy image pairs, ACM Transactions on Graphics 26(3), 2007, article 1, DOI: 10.1145/1276377.1276379.
• [24] SHAN Q., JIA J., AGARWALA A., High-quality motion deblurring from a single image, ACM Transactions on Graphics 27(3), 2008, article 73, DOI: 10.1145/1360612.1360672.
• [25] CHAN S.H., WANG X., ELGENDY O.A., Plug-and-play ADMM for image restoration: fixed-point convergence and applications, IEEE Transactions on Computational Imaging 3(1), 2017, pp. 84–98, DOI: 10.1109/TCI.2016.2629286.
• [26] MOSLEH A., SOLA Y.E., ZARGARI F., ONZON E., LANGLOIS J.M.P., Explicit ringing removal in image deblurring, IEEE Transactions on Image Processing 27(2), 2018, pp. 580–593, DOI: 10.1109/TIP.2017.2764625.
• [27] CAO S., HE N., ZHAO S., LU K., ZHOU X., Single image motion deblurring with reduced ringing effects using variational Bayesian estimation, Signal Processing 148, 2018, pp. 260–271, DOI: 10.1016/j.sigpro.2018.02.015.
• [28] SHAO W.-Z., DENG H.-S., GE Q., LI H.-B., WEI Z.-H., Regularized motion blur-kernel estimation with adaptive sparse image prior learning, Pattern Recognition 51, 2016, pp. 402–424, DOI: 10.1016/j.patcog.2015.09.034.
• [29] ZHOU N.R., LUO A.W., ZOU W.P., Secure and robust watermark scheme based on multiple transforms and particle swarm optimization algorithm, Multimedia Tools and Applications 78(2), 2019, pp. 2507 –2523, DOI: 10.1007/s11042-018-6322-9.
• [30] ABRAMOVICH F., RITOV Y., Statistical Theory: A Concise Introduction, CRC Press, Florida, USA 2013, pp. 121–129.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).