摘要
为了提高太赫兹图像的质量,克服边缘模糊的缺陷,采用有理分形插值和基于梯度变换的图像超分辨率重建算法相结合的方法对0.25THz,0.50THz和0.75THz图像进行超分辨率重建实验,并对实验结果进行了定量分析,利用基于空间信息熵的直方图匹配技术和双边滤波器对重建算法进行了优化,增强了该方法的适用性。结果表明,对经过插值的太赫兹图像采用基于梯度变换的超分辨率重建方法处理之后,0.25THz,0.50THz和0.75THz图像的边缘强度分别提高了169%,116%和104%,平均梯度分别提高了16%,28%和24%;同时,成像信号频率和强度也会对重建性能产生影响。该方法可以有效恢复太赫兹图像当中的细节信息,锐化图像边缘,提高图像质量且不会出现振铃现象,具有较好的实用价值。
In order to improve the quality of terahertz image and overcome the problem of edge blur of terahertz image,a super-resolution reconstruction method,which combines rational fractal interpolation and gradient field transform,was proposed in this paper for terahertz image reconstruction with frequencies of 0.25THz,0.50THz,and 0.75THz.Meanwhile,spatial entropy-based image enhancement and bilateral filtering are introduced to optimize the reconstruction.Experimental results illustrate that after processing interpolated terahertz images by super-resolution reconstruction based on gradient field transform,the edge strength of images with frequencies of 0.25THz,0.50THz,and 0.75THz were respectively improved by 169%,116%,and 104%,and the average gradient of those images were improved by 16%,28%,and 24%,respectively.Moreover,signal frequency and intensity would affect the performance of the reconstruction.This method can recover the detail information in terahertz images,sharpen edges of the object,improve the quality of terahertz images without ringing effect,and has practical value.
作者
郭佑东
凌福日
姚建铨
GUO Youdong;LING Furi;YAO Jianquan(Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan 430074, China;School of Optical and Electronic Information, Huazhong University of Science and Technology, Wuhan 430074, China;College of Precision Instrument and Opto-electronics Engineering, Tianjin University, Tianjin 300072, China)
出处
《激光技术》
CAS
CSCD
北大核心
2020年第3期271-277,共7页
Laser Technology
基金
国家重点基础研发计划资助项目(2015CB755403)。
关键词
图像处理
太赫兹
超分辨率重建
插值
梯度变换
边缘锐化
image processing
terahertz
super-resolution
interpolation
gradient field transformation
edge sharpening
