摘要
针对图像增强过程中出现的清晰度低、噪声放大、细节不明显和边缘梯度保持能力弱等问题,提出一种NSST和引导滤波相结合的图像增强算法。将原始图像和直方图均衡化后的图像分别进行NSST处理,得到各自的低频子带和高频子带;将直方图均衡化后的低频子带进行PM滤波增强;采用循环阈值法抑制原始图像的高频子带中的噪声,用引导滤波对高频子带进行增强,提高图像的清晰度和边缘梯度保持能力,对处理后的低频子带和高频子带进行NSST逆变换。实验结果表明,该算法增强了图像的细节信息,改善了图像的视觉效果。
Aiming at the problems of low resolution,amplifying noise,lack of the details and weak edge gradient retention in the process of the image enhancement,an image enhancement algorithm combining NSST and guided filtering was proposed.The image of the original image and the histogram equalization were divided into NSST decomposition,and the respective low-frequency sub-bands and the high frequency sub-bands were obtained.The low frequency subband of histogram equalization was enhanced by PM filtering.The NSST with cycle spinning approach was used to suppress the high frequency sub-bands noise of the original image,the high frequency sub-bands were enhanced using guide filter to improve the definition and edge gradient ability of the images.The reconstructed image was obtained through inverse NSST for the processed low-frequency and high frequency sub-bands.Experimental results show that the algorithm can enhance image details,the visual effect of the image is greatly improved.
作者
韩晶
贾振红
杨杰
Nikola Kasabov
HAN Jing;JIA Zhen-hong;YANG Jie;Nikola Kasabov(College of Information Science and Engineering,Xinjiang University,Urumqi 830046,China;Institute of Image Processing and Pattern Recognition,Shanghai Jiao Tong University,Shanghai 200240,China;Knowledge Engineering and Discovery Research Institute,Auckland University of Technology,Auckland 1020,New Zealand)
出处
《计算机工程与设计》
北大核心
2018年第9期2832-2835,2906,共5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(2014-2029
2016-2196)
关键词
遥感图像
图像增强
NSST变换
PM滤波
旋转阈值去噪
引导滤波
remote sensing image
image enhancement
NSST transformation
PM filtering
NSST with cycle spinning approach
guided filtering
