计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 231200046-9.doi: 10.11896/jsjkx.231200046

• 图像处理&多媒体技术 • 上一篇    下一篇

基于频率域点扩散函数的视频中运动模糊车牌复原与识别优化

朱沛伍, 高树辉, 谢昭宇, 傅裕   

  1. 中国人民公安大学侦查学院 北京 100038
  • 出版日期:2024-11-16 发布日期:2024-11-13
  • 通讯作者: 高树辉(gaoshuhui@ppsuc.edu.cn)
  • 作者简介:(281228621@qq.com)
  • 基金资助:
    中国人民公安大学刑事科学技术双一流创新研究专项(2023SYL06)

Optimization of License Plate Image Restoration and Recognition in Video with Motion BlurBased on Frequency Domain Point Spread Function

ZHU Peiwu, GAO Shuhui, XIE Zhaoyu, FU Yu   

  1. School of Investigation,People's Public Security University of China,Beijing 100038,China
  • Online:2024-11-16 Published:2024-11-13
  • About author:ZHU Peiwu,born in 2000,postgra-duate.His main research interest is forensic science of image.
    GAO Shuhui,born in 1971,Ph.D,professor,Ph.D supervisor.Her main research interest is forensic science of image.
  • Supported by:
    Double First-class Innovation Research Project of Forensic Science,People's Public Security University of China(2023SYL06).

PDF (PC)

150

摘要： 车辆是刑事案件中常使用的工具,车牌识别是涉案车辆认定的重要依据之一,因此对运动模糊造成的车牌图像退化的复原是刑事技术工作的重要研究方向。文中提出一种基于频率域点扩散函数参数估计的车牌图像复原算法,运用二维离散傅里叶变换、Radon变换以及Wiener滤波算法解决监控视频视角下运动模糊车牌的复原问题。通过预处理对运动模糊图像进行灰度化和降噪,并对图像添加Hanning窗,进行二维离散傅里叶变换和对数运算,计算图像的功率频谱图,运用Radon变换对频谱图进行检测,估算模糊方向;对频谱图进行最小值计算,求模糊长度,完成对点扩散函数两个参数的估算。通过Wiener滤波算法进行图像反卷积,实现图像复原。针对传统频谱图估算模糊参数易受中央亮线干扰的问题,在离散傅里叶变换之前添加窗函数进行优化。使用道路监控拍摄的模糊图像,进行点扩散函数参数估计实验,同时与目前未加窗的方法进行对比,验证所提算法的复原效果。实验结果证明,所提算法在复原车牌图像语义信息方面具有优势,能为刑事图像技术领域模糊车牌复原提供方法补充。

关键词: 点扩散函数, PSF, 图像复原, 运动视频, 模糊车牌, 图像处理

Abstract: Vehicles are commonly used tools in criminal cases,and license plate recognition is one of the crucial criteria for identi-fying involved vehicles.The restoration of license plate images degraded by motion blur is an important research direction in digi-tal image processing.This paper proposes a license plate image restoration algorithm based on the estimation of frequency domain point spread function parameters.It utilizes two-dimensional discrete Fourier transformation,Radon transformation,and the Wiener filtering algorithm to address the problem of restoring license plates affected by motion blur in surveillance video angles.The process starts with preprocessing the motion-blurred image by converting it to grayscale and reducing noise.A Hanning window is applied to the image,followed by two-dimensional discrete Fourier transformation and logarithmic operations.This computes the power spectrum of the image.Radon transformation is used to detect the spectrum and estimate the blur direction.The minimum value computation on the spectrum determines the blur length,which completes the estimation of the two parameters of the point spread function.The image is then deconvolved using the Wiener filtering algorithm,resulting in the restoration of the image.To address the issue of traditional spectrum estimation being susceptible to interference from central bright lines,a window function is added before the discrete Fourier transformation.To validate the algorithm's restoration effect,experiments are conducted with motion-blurred images captured by road surveillance.These experiments are compared with the non-windowed me-thod,establishing a method for restoring motion-blurred license plate images.Experimental results demonstrate that the proposed algorithm has an advantage in preserving the semantic information of license plate images and can complement the field of criminal image technology in the restoration of blurred license plates.

Key words: Point spread function, PSF, Image restoration, Motion video, Blurred license plate, Image processing

中图分类号: 

  • TP309
[1]ZHANG J H,DU J,ZHU Y D,et al.A Fast Point Spread Function Parameter Extraction Algorithm Based on Linear Regression [J].Optical Technique,2023,49(5):534-539.
[2]HOU W H,WU Y J.Evaluating the resolution of conventional optical microscopes through point spread function measurement[J].iScience,2023,26(10):107976.
[3]CANNON M.Blind deconvolution of spatially invariant image blurs with phase[J].Acoustics Speech & Signal Processing IEEE Transactions on,1976(1):58-63.
[4]LU Q B,ZHOU W G,FANG L,et al.Robust Blur Kernel Estimation for License Plate Images From Fast Moving Vehicles[J].IEEE Transactions on Image Processing:a Publication of the IEEE Signal Processing Society,2016,25(5):2311-2323.
[5]MAMTA R,DUTTA M.GA based Blind Deconvolution Technique of Image Restoration using Cepstrum Domain of Motion Blur[J].Indian Journal of Science and Technology,2017,10(16):1-8.
[6]KANG X,PENG Q,THOMAS G,et al.Blind Image Restoration using the Cepstrum Method[C]//Canadian Conference on Electrical and Computer Engineering,2006(CCECE 06).IEEE,2007.
[7]XU X Y,PAN J S,ZHANG Y J,et al.Motion blur kernel estimation via deep learning[J].IEEE Transactions on Image Processing,2018,27(1):194-205.
[8]SCHULERC J,HIRSCH M,HARMELING S,et al.Learning to Deblur[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2016,38(7):1439-1451.
[9]WANG X H,ZHAO R C.Elimination of motion blur in any direction [J].Journal of Image and Graphics,2000,5(6):525-529.
[10]WANG X H,ZHAO R C.Estimation of PSF for Uniform Linear Motion Blurring [J].Journal of Computer Applications,2001,21(9):40-41.
[11]CHEN J,ZHANG X,CHEN Z R.Improved PSF parameter estimation algorithm for motion blurred images based on Radon transform [J].Computer Engineering & Software,2020,41(6):1-6.
[12]TANG C J.Parameter detection of motion blurred images based on spectral analysis [J].Information and Electronic Engineering,2015,13(1):148-153.
[13]ZHANG G Y.Research on Global Uniform Linear MotionBlurred Image Restoration Algorithm [D].Wuahan:Hubei University of Technology,2021.
[14]SUN P,LANG Y B,MUBALAK A,et al.Motion Blurred Li-cense Plate Image Restoration Method Based on PSF Parameter Estimation [J].Journal of Criminal Investigation Police University of China,2017(4):121-124.
[15]XU H S.Parameter estimation and restoration of motion blurred license plate images in surveillance videos [J].Journal of Hebei Vocational College of Public Security Police,2021,21(2):28-32.
[16]LE X,CHENG J,LI M.An improved motion blur image parameter estimation method based on Radon transform [J].Infrared and Laser Engineering,2011,40(5):963-969.
[17]JU S Y,GAO S H.A method for improving the accuracy of PSF parameters in motion blurred images using window functions [J/OL].Laser & Optoelectronics Progress:1-13.[2022-10-12].http://kns.cnki.net/kcms/detail/31.1690.TN.20230207.1646.076.html.
[18]BUCKER A S,STEVEN S,DAVID V,et al.Deep learning estimation of modified Zernike coefficients and recovery of point spread functions in turbulence[J].Optics Express,2023,31(14):22903-22913.
[19]YOGESH K,SHEFALI S,MAYANK S,et al.Characterization of instrumental PSF in neutron imaging experiments using logarithmic power spectral plot method[J].NDT and E Interna-tional,2023,139:102922.
[20]MING W H.Research on Motion Blurred Image Restoration Algorithm [D].Hefei:Anhui University,2004.
[21]JIA H T,ZHU Y C,WANG J H.Principle and Implementation of Extended Adaptive Median Filter [J].Journal of Image and Graphics,2004(8):55-58.
[22]CHEN Q L,YIN Y Y,JIANG Q,et al.High resolution hyper lens imaging based on SRResNet and point spread function design [J].Optical Technique,2023,49(4):385-389,411.
[23]FENG R.Research on Key Algorithms for Image Clarity in Surveillance Video [D].Shenyang:Northeastern University,2016.
[24]FU J Q.Research and Application of License Plate Recognition Algorithm in Complex Background [D].Shanghai:Fudan University,2015.
[25]XU K.Motion Blurred Image Restoration in Low Light Scenarios [D].Wuhan:Central China Normal University,2019.
[26]ZHANG M.Super resolution motion blurred image restoration based on Radon transform [D].Chengdu:Southwest Jiaotong University,2012.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发