摘要
针对MAP(maximum a posteriori)框架下QR码(quick response code)去模糊方法复原效果不佳、运行时间长的问题,为提高模糊QR码的解码能力,提出一种QR码快速去模糊方法。在改进的MAP框架下引入基于灰度分布特性的图像先验,约束复原图像的二值特性,可有效提高QR码的复原效果,并采用改进的多尺度模糊核估计方法,抑制噪声干扰。实验表明,相比于其他基于MAP框架的二值图像去模糊方法,该方法在复原效果、运行时间以及复原后图像的识别率上均有明显优势。
To address the issues of poor restoration effects and long running times in existing maximum a posteriori(MAP)-based deblurring methods for quick response(QR)codes,a fast QR code deblurring method is proposed to enhance the decoding capability of blurred QR codes.Within the improved MAP framework,an image prior based on grayscale distribution was introduced to constrain the binary characteristics of the QR code and improve the image quality.Additionally,an improved multi-scale blur kernel estimation method was employed to suppress image noise.Experiments demonstrate that,compared to other MAP-based deblurring methods for binary images,this method exhibits considerable improvements in both deblurring effect and computational efficiency.
作者
施逸凡
陈宁
SHI Yifan;CHEN Ning(College of Marine Equipment and Mechanical Engineering,Jimei University,Xiamen 361021,China)
出处
《集美大学学报(自然科学版)》
2025年第4期402-408,共7页
Journal of Jimei University:Natural Science
基金
福建省自然科学基金项目(2021J01851)。
关键词
QR码
去模糊
灰度分布
MAP框架
QR(quick response)code
deblurring
grayscale distribution
MAP(maximum a posteriori)
