摘要
本文主要研究基于灰度共生矩阵的纹理分析方法,提取医学PET图像肿瘤靶区的纹理特征,用于量化恶性肿瘤的异质性。Haralick等首先提出了灰度共生矩阵的概念,Veenland等应用纹理特征对图像进行分析。本论文基于灰度共生矩阵对医学PET图像进行纹理分析,通过算法计算得到肿瘤图像数字化信息,不仅丰富了图像信息,更利于分析恶性肿瘤的变化情况。
In this paper,we mainly study the texture analysis method based on gray level co-occurrence matrix,and extract the texture features of tumor target in medical PET image.Texture features are used to quantify the heterogeneity of malignant tumor.Haralick et al first proposed the concept of gray co-occurrence matrix,Veenland et al applied texture features to analyze the image.In this paper,the texture of medical PET images is analyzed based on the gray covariance matrix.The digital information of tumor images is calculated by the algorithm,which not only enriches the image information,but also analyzes the changes of malignant tumors.
出处
《电脑知识与技术（过刊）》
2017年第2X期219-220,共2页
Computer Knowledge and Technology
关键词
PET图像
纹理分析
灰度共生矩阵
PET Image
Texture Analysis
Gray Level Co-occurrence Matrix
