摘要
对图像纹理特征进行了研究,采用改进的共生矩阵算法来提取纹理特征.在此基础上,通过高斯归一化法用纹理特征作为图像检索特征,给出了利用该特征的图像检索算法.从实验结果可以看出,对于具有明显的纹理特征的图像数据库利用改进的共生矩阵提取算法具有较好的检索性能,可以得到较好的检索效果.
In this paper, improved co-occurrence matrix algorithm is applied to image texture extraction. On this basis, along with texture processed by normalized Gaussian function, image retrieval algorithm is presented. As can be seen from the results, better retrieval performance is obtained with improved co-occurrence matrix extraction algorithm.
出处
《青海师范大学学报（自然科学版）》
2012年第4期20-24,共5页
Journal of Qinghai Normal University(Natural Science Edition)
关键词
图像检索
特征提取
相似性度量
共生矩阵
content-based image retrieval
feature extraction
similarity measurement
co -- occurrence matrix
