|本期目录/Table of Contents|

[1]王锦红,陈文△.基于小波变换的异常特征超声图像定位技术*[J].生物医学工程研究,2019,02:176-180.
 WANG Jinhong,CHEN Wen.Ultrasonic image localization of abnormal features based on wavelet transform[J].Journal of Biomedical Engineering Research,2019,02:176-180.
点击复制

基于小波变换的异常特征超声图像定位技术*(PDF)

《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]

期数:
2019年02期
页码:
176-180
栏目:
出版日期:
2019-06-25

文章信息/Info

Title:
Ultrasonic image localization of abnormal features based on wavelet transform
文章编号:
1672-6278 (2019)02-0176-05
作者:
王锦红1陈文2△
1.四川省绵阳市中心医院,四川 绵阳 621000;2.西南医科大学,四川 泸州 646000
Author(s):
WANG JinhongCHEN Wen
1.Mianyang Central Hospital of Sichuan,Mianyang 621000,China;2.Southwest Medical University ,Luzhou 646000,China
关键词:
小波变换异常特征超声图像诊断识别边缘轮廓检测梯度分解滤波
Keywords:
Wavelet transformAbnormal feature Ultrasound imageDiagnosis and recognitionEdge profile detectionGradient decompositionFiltering
分类号:
R318;TP399
DOI:
10.19529/j.cnki.1672-6278.2019.02.10
文献标识码:
A
摘要:
为了提高超声图像的诊断识别能力,需要进行异常特征点定位。提出一种基于小波变换的异常特征超声图像定位技术。对采集的超声图像采用边界特征融合方法进行边缘轮廓检测,在邻域内采用颜色梯度分解方法进行超声图像区域融合滤波处理,结合小波变换方法进行超声图像的特征分解和尺度模板匹配,提取超声图像的奇异特征点,根据超声图像的颜色特征分解值进行邻域均衡控制和自适应特征参量估计,根据超声图像的纹理和颜色特征提取结果进行图像的异常特征点定位检测和识别。仿真结果表明,采用该方法进行超声图像异常特征点定位的准确度较高,定位精度较好,提高了超声图像的异常特征辨识能力。
Abstract:
In order to improve the ability of ultrasonic image diagnosis and recognition, it is necessary to locate abnormal feature points. A new ultrasonic image localization technique based on wavelet transform was proposed. The boundary feature fusion method was used to detect the edge contour of the collected ultrasonic image, and the color gradient decomposition method was used to process the ultrasonic image region fusion filter in the neighborhood. The wavelet transform method was used to decompose the ultrasonic image and match the scale template, extract the singular feature points of the ultrasonic image, control the neighborhood equalization and estimate the adaptive characteristic parameter according to the color decomposition value of the ultrasonic image. Based on the texture and color features of ultrasonic images, the abnormal feature points were detected and recognized. The simulation results show that the method has high accuracy and good localization accuracy for abnormal feature points in ultrasonic images, and improves the ability of identifying abnormal features in ultrasonic images.

参考文献/References

备注/Memo

备注/Memo:
(收稿日期:2018-12-22)四川省卫计委基金资助项目(2016SZ169)。△通信作者Email:308675935@qq.com
更新日期/Last Update: 2019-07-18