摘要
针对遥感图像的细节信息丰富、对比度低、清晰度差等特点,提出基于剪切波变换和反锐化掩膜相结合的遥感图像增强算法。直方图均衡化遥感图像,剪切波分解遥感图像,得到低频子带系数和高频子带系数,对各尺度各方向的高频系数进行自适应阈值去噪和线性增强处理。为防止过增强,不处理剪切波变换产生的低频系数,对做逆变换回空间域的图像采用反锐化掩膜来增强边缘和细节。实验结果表明,该算法对遥感图像有较好的增强效果。
Aiming at the characteristics of remote sensing image such as rich details,low-contrast and poor definition,an enhancement algorithm for remote sensing image based on Shearlet transform and unsharp masking was presented.First,the histogram equalization was carried out on remote sensing image.Second,the remote sensing image was decomposed into low frequency coefficients and high frequency coefficients using Shearlet transform,and the low frequency subband part was not processed,adaptive threshold denoising and linear enhancement were used on the high frequency coefficients of each scale and direction.At last the remote sensing image was transformed into frequency domain and the unsharp masking was used to enhance the details and edges.The experiment results demonstrate that the proposed method has significant effects on remote sensing image enhancement.
出处
《计算机工程与设计》
北大核心
2015年第4期987-990,共4页
Computer Engineering and Design
基金
教育部促进与美大地区科研合作与高层次人才培养基金项目(20101595)
关键词
遥感图像
直方图均衡化
剪切波变换
自适应阈值去噪
反锐化掩膜
remote sensing image histogram equalization Shearlet transform adaptive threshold denoising unsharp masking
