节点文献
综合纹理特征和语义信息的医学图像检索
Medical image retrieval based on texture features and semantic information
【摘要】 提出一种结合图像分块纹理特征和语义信息的医学胸片图像检索方法。同时,介绍了颜色特征提取方法中的颜色相关图算法。据此,实现了一个图像检索原型系统,依据所设计的评价实验,将不同实验的检索结果进行了比较和分析。实验证明,结合图像分块纹理特征和语义信息的检索方法具有较好的检索效果。
【Abstract】 This paper presented a method of medical images retrieval about sternums based on texture features combining with semantic information.The texture features were calculated with gray level co-occurrence matrix in different partitions.This paper also introduced an algorithm about color correlogram.A prototype system was implemented.Meanwhile,three experiments were designed to evaluate the effectiveness of the approach.The comparison of the retrieval results show that the approach proposed in the paper is effective.
【关键词】 医学图像检索；
灰度层共现矩阵；
颜色相关图；
语义信息；
【Key words】 medical image retrieval； gray level co-occurrence matrix； color correlogram； semantic information；
【Key words】 medical image retrieval； gray level co-occurrence matrix； color correlogram； semantic information；
【基金】 国家自然科学基金资助项目(60472099)
- 【文献出处】 计算机应用 ,Computer Applications , 编辑部邮箱 ,2005年10期
- 【分类号】TP391.3
- 【被引频次】5
- 【下载频次】258