基于血液斑点噪声抑制和T-Snake模型的血管内超声图像边缘提取

宫延新; 孙丰荣; 刘泽; 郑晓势; 赵大哲; 刘积仁

doi:10.11834/jig.20070411

图像分析和识别 | 浏览量 : 0 下载量: 202 CSCD: 0

PDF
导出
分享
收藏
专辑

基于血液斑点噪声抑制和T-Snake模型的血管内超声图像边缘提取
Edge Detection of IVUS Images Based on Blood Speckle Noise Reduction and the T-Snake Model
2007年12卷第4期页码：655
纸质出版：2007
DOI： 10.11834/jig.20070411
稿件说明：

移动端阅览

宫延新, 孙丰荣, 刘泽, 郑晓势, 赵大哲, 刘积仁. 基于血液斑点噪声抑制和T-Snake模型的血管内超声图像边缘提取[J]. 中国图象图形学报, 2007,12(4):655. DOI： 10.11834/jig.20070411.

Edge Detection of IVUS Images Based on Blood Speckle Noise Reduction and the T-Snake Model[J]. Journal of Image and Graphics, 2007, 12(4): 655. DOI： 10.11834/jig.20070411.

摘要

血管内超声(IVUS)图像冠状动脉血管壁内、外膜的边缘提取对冠状动脉疾病的诊断和治疗有着重要意义。为了更好地抑制血管内超声图像的血液斑点噪声

首先采用一种时/空滤波方法对IVUS图像进行降噪预处理

以抑制其严重的血液斑点噪声;然后

为了更好地提取血管边缘

提出了一种改进的自适应形变模型

并基于该改进的自适应形变模型(T-Snake模型)给出了一种IVUS图像冠状动脉血管壁内、外膜边缘的提取方法。实验结果表明

该边缘提取方法有着较高的准确性和可靠性

对IVUS序列图像处理的可重复性和鲁棒性较好;也表明了改进的T-Snake模型的可实现性

以及IVUS图像血液斑点噪声抑制方法的有效性。

Abstract

Edge detection of intravascular ultrasound(IVUS) images is important for the diagnosis and treatment of the coronary artery disease.We adopt a spatial/temporal adaptive filtering method to reduce noise in IVUS images which contain severe blood speckle noise.An improved topologically adaptable Snakes madel(T-Snake) is proposed in the paper

based on which a method for automatically detecting the edge of IVUS image is presented.Experiments show that the edge detection method is accurate

reproducible and robust for sequential IVUS frames and also the improved T-Snake is effective and realizable and the presented method for reducing the blood speckle noise is effective as well.