摘要
为提高医学超声图像的边缘识别水平,提出一种基于相似性度量方法的加权医学超声图像对比度增强方法。通过相似性度量方法计算围绕某一像素点的处理窗口与匹配模板的相似值,根据相似值与超声图像各部分特征的关系,对达到一定相似程度的已匹配图像通过局部增强算法进行增强,最后通过多图像加权平均法获得最终增强结果。实验结果表明,该方法不仅能有效抑制噪声,而且能准确检测出超声各特性区域图像,在增强边缘效果的同时也能较为完整的保留原有图像的细节信息,弥补了传统增强算法不能兼顾对比度增强和保持图像细节的不足。
To improve edge detection level of medical ultrasonic images,a weighted contrast enhancement method of medical ultrasound images based on similarity estimation is proposed.Firstly the similarity of the processing window around a pixel and matching template is calculated.According to the relationship between the similarity and each parts of ultrasound image,the matched images are enhanced by local enhancement approach when it satisfied with correlation condition.The weighted average method for multiple images is used to obtain the final enhancement results.Experimental results with this method shows better contrast resolution,better edge feature detection than the original,and also preserved the structure without losing useful clinical information.In this respect,this method remedies the shortage of traditional enhancement algorithms.
出处
《计算机工程与设计》
CSCD
北大核心
2011年第3期995-997,1001,共4页
Computer Engineering and Design
基金
四川省教育厅自然科学重点基金项目(08ZA005)
关键词
相似度
边缘增强
医学超声图像
直方图均衡化
归一化互相关
绝对误差和
similarity degree
edge enhancement
medical ultrasonic images
histogram equalization
normalized cross correlation
sum of absolute difference
