乳腺肿瘤图像的融合纹理特征提取方法  

Extraction of Fusion Feature from Breast Cancer Images

作　　者：汪友明[1] 张菡玫 WANG Youming;ZHANG Hanmei(Xi'an University of Posts and Telecommunications,Xi'an 710121)

机构地区：[1]西安邮电大学

出　　处：《计算机与数字工程》2019年第6期1497-1501,共5页Computer & Digital Engineering

摘　　要：为了准确地识别乳腺肿瘤图像特征,提出一种改进的灰度共生矩阵(GLCM)与Tamura相结合的纹理特征提取算法。首先对乳腺肿瘤图像进行预处理,滤除图像噪声部分并提高图像的对比度;通过对传统的灰度共生矩阵进行改进,减少大量的冗余信息,增强图像的识别率;最后将改进的灰度共生矩阵与Tamura相结合,提取图像纹理特征,并进行图像特征识别。实验结果表明,乳腺肿瘤图像特征识别率可达到96.67%,平均计算时间为14.6s,具有较高的识别准确率和计算效率。In order to accurately identify breast tumor image features,an improved textural feature extraction algorithm based on improved gray level co-occurrence matrix(GLCM)and Tamura is proposed. First,the images are preprocessed by eliminating image noise and enhancing image contrast. Then,the traditional gray level co-occurrence matrix is improved,and redundant information is reduced,improving the recognition rate of the image and the running speed of the program. Finally,The symbiotic matrix is combined with Tamura to obtain the image texture features,and the extracted features are identified. The experimental results show that the recognition rate of fusion features can reach 96.67% with 14.6 s average calculation rate,which has high recognition accuracy and omputational efficiency.

关 键 词：纹理特征 灰度共生矩阵 Tamura纹理 图像分类 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]