摘要
针对X射线数字实时成像对比度低、受材料及厚度影响大等特点,提出一种基于双能和小波融合技术的X射线图像增强算法,以解决单能谱条件下细节丢失的问题.结合双能理论,对基于小波的融合规则进行相应改进,低频分量采取分形维数最大熵线性加权的融合规则,高频分量采取梯度值取大再乘以调整系数的融合规则,通过小波重构获得融合图像.将所得融合图像与原始图像和对比融合图像进行比较,并采用平均梯度、标准差、信息熵以及空间频率等客观参数进行定量评价.结果表明,此方法对单能谱条件下细节丢失的图像有较好的图像增强效果,且不会产生双能图像互补信息丢失.
Regarding the low contrast of X-ray digital radiography images and large influence from materi- als and thickness, an X-ray image enhancement algorithm was proposed. The algorithm was based on du- al-energy and wavelet fusion to solve the problem of detail loss in single-energy imaging. With dual-ener- gy theory, improvement of fusion rules was made based on wavelet. Sub-image with low frequency was carried out with a fractal dimension and maximal entropy and linear weighted fusion rule, and sub-image with high frequency was carried out with a larger gradient and weighted fusion rule, and then fusion image was obtained through wavelet reconstruction. The fusion image was compared with original images and fu- sion images with contrast fusion methods for qualitative analysis, while they were quantificationally evalu- ated by using average gradient, standard deviation, information entropy and spatial frequency. Result shows an excellent image enhancement effect for detail loss in single-energy imaging and information is not lost in fusion image.
作者
胡春光
靳丽媛
邹晶
胡晓东
Hu Chunguang Jin Liyuan Zou Jing Hu Xiaodong(School of Preeision Instrument and Opto-Eleetronics Engineering, Tianjin University, Tianjin 300072, China Center for Mathematics and Information Interdisciplinary Science （SCMIIS） , Beijing 100000, China)
出处
《纳米技术与精密工程》
CAS
CSCD
北大核心
2016年第6期429-433,共5页
Nanotechnology and Precision Engineering
基金
国家自然科学基金资助项目(61373144)
关键词
X射线成像
双能
小波变换
图像融合
X-ray imaging
dual-energy
wavelet transform
image fusion
