A New Parametric Kernel Estimation Technique for License Plate Image De-blurring
- 151 Downloads
A recognizable license plate in a picture taken by a traffic monitoring system is crucial for identifying the vehicles involved in traffic violations. In the image of a vehicle taken by a surveillance camera, the license plate is often blurred due to fast motion and cannot be recognized by the human eye. In this type of blurring, the blur kernel can be seen to be a linear uniform convolution parametrically described by its angle and length. In this paper, we introduce a new estimation technique to determine this kernel accurately in order to improve our de-blurred result. We use the Hough transform in estimating the direction in which the image is blurred. To determine the extent of the blur in that direction, we employ a new method involving the cepstrum of the blurred image. We compare the performance of our method to that of other recent blind de-blurring techniques. These comparisons show that our proposed scheme can handle significant blur in the captured image to give a good output image.
KeywordsMotion kernel Blur angle Blur length Hough transform Cepstral transform Point spread function Optical transfer function Ground truth Blind image de-blurring
- 2.Zhou, W., Lu, Y., Li, H., Song, Y., & Tian, Q. (2016). Spatial coding for large scale partial-duplicate Web image search. In Proceedings of the 18th ACM International Conference on Multimedia (pp. 511–520).Google Scholar
- 7.Xu, L., Zheng, S. & Jia, J. (2013). Unnatural sparse representation for natural image de-blurring. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1107–1114).Google Scholar
- 8.Cho, H., Wang, J., & Lee, S. (2012). Text image de-blurring using text-specific properties. In Proceedings of the European conference on computer vision (pp. 524–537).Google Scholar
- 9.Xu, L., & Jia, J. (2010). Two-phase kernel estimation for robust motion de-blurring. In Proceedings of the European conference on computer vision (pp. 157–170).Google Scholar
- 15.Gupta, A., Joshi, N., Zitnick, C. L., Cohen, M., & Curless, B. (2010) Single image deblurring using motion density functions. In Proceedings of the 11th European conference on computer vision (pp. 171–184).Google Scholar
- 16.Zheng, S., Xu, L., & Jia, J. (2013). Forward motion deblurring. In Proceedings of the IEEE international conference on computer vision (pp. 1465–1472).Google Scholar
- 17.Tiwari, S., Shukla, V. P., Singh, A. K., & Biradar, S. R. (2013). Review of motion blur estimation techniques. Journal of Image and Graphics, 1(4), 176–184.Google Scholar
- 18.Gonzalez, R. C., & Woods, R. E. (2007). Digital Image Processing. Englewood Cliffs: Prentice Hall.Google Scholar
- 20.Krishnan, D., Tay, T., & Fergus, R. (2011). Blind deconvolution using a normalized sparsity measure. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 233–240).Google Scholar
- 22.Chang, C. C., & Lin, C. J. (2016). A library for support vector machines. Available at: http://www.csie.ntu.edu.tw/~cjlin/libsvm.
- 23.Oliveira, J. P., Figueiredo, M. A. T., & Bioucas-Dias, J. M. (2007). Blind estimation of motion blur parameters for image deconvolution. In Proceedings of the 3rd Iberian conference on pattern recognition and image analysis (pp. 604–611).Google Scholar