您好,欢迎来到中国测试科技资讯平台!

首页> 《中国测试》期刊 >本期导读>超声图像中目标的边界提取

超声图像中目标的边界提取

2095    2016-01-23

免费

全文售价

作者:刘艳丽, 刘奇

作者单位:四川大学电气信息学院, 四川成都 610065


关键词:超声图像; Gabor滤波器; 形态学操作; 尺度参数; 边界提取


摘要:

由于对比度低和噪声强等特点,超声图像分割很难达到满意的效果,针对这一状况,将Gabor滤波器结合形态学操作应用到超声图像中目标的边界提取。首先对原始图像做预处理,去噪增强,然后再利用Gabor滤波进行目标区域特征提取,再经形态学腐蚀、膨胀等操作平滑图像,最后由sobel算子提取最终的边界图像。该法的关键之处在于Gabor滤波器尺度参数的选择,通过实验和分析得出了一组最佳尺度参数,取得了良好的分割效果,同时也验证了这一方法在超声图像分割中的实用性。


Target edge extraction for ultrasonic images

LIU Yan-li, LIU Qi

School of Electrical Engineering and Information, Sichuan University, Chengdu 610065, China

Abstract: Due to poor contrast and severe speckle noise, it is difficult to obtain satisfy segmentation for ultrasonic images. To solve this problem, the target edges are extracted for ultrasonic images with Gabor filter combined with morphological operation. Firstly, the original image is preprocessed for the purpose of noise suppression and image enhancement, then the Gabor filter is applied to extract the feature of the region connected with the border of the targets, after the region has been smoothed by morphology operation, sobel operation is chosen to acquire the final borders of the targets. The key point of this method is to set the scale parameters of Gabor filter. Through the experiments and analysis, a group of the best scale parameters has been obtained, which makes a good segmentation. At the same time, the practicality of the method has also been verified for ultrasonic image segmentation.

Keywords: Ultrasonic images; Gabor filter; Morphology operation; Scale parameter; Edge extraction

2008, 34(6): 66-68  收稿日期: 2008-5-1;收到修改稿日期: 2008-7-12

基金项目: 

作者简介: 刘艳丽(1983-),女,河北承德市人,硕士研究生,专业方向为信号与信息处理。

参考文献

[1] Jain A K,Farrokhnia F.Unsupervised texture segmentation using gabor filters[J]. Pattern Recognition,1991,24(12):1167-1186.
[2] Nikhil R P,Sanker K P. A review on image segmentation techniques[J]. Pattern Recognition,1993,26(9):1277-1294.
[3] 陈静,罗斌.基于Gabor滤波的指纹图像增强算法[J].计算机技术与发展,2008.
[4] Wu X, Bhanu B. Gabor wavelet representation for 3-D object recognition[J]. IEEE Trans. Image Processing, 1997,9(1):47-64.
[5] Raghu P P,Poongodi R,Yegnanarayan B. Unsupervised texture classification using vector quatization and deterministic relaxation on neural network[J]. IEEE Trans. Image Processing.Oct,1997,6(10):1376-1387.
[6] 李宏贵,李兴国.基于Gabor小波滤波器的红外图像多尺度识别[J].红外与毫米波学报,2000.
[7] Mehrotora R,Gabor. Filter based edge detection[J]. Pattern Recogniation,1992,25(12):1479-1494.