摘要
将图像分割算法和模糊理论相结合,提出了基于区域模糊直方图的图像检索方法。用快速聚类算法对图像进行区域分块,以各区域的视觉特征为基础,建立图像模糊直方图并进行图像检索。在1000幅通用图像库上的实验表明,该方法可以得到比较满意的检索结果。
An image retrieval method based on the region fuzzy histogram is proposed by combining the image segmentation algorithm and the fuzzy theory. First, the k-means clustering algorithm is used to segment the images and the fuzzy histogram is built for the image retrieval based on the visual features of every region. Experiment results from the database with 1000 general images show that this method is satisfactory for the image retrieval.
出处
《江苏科技大学学报（自然科学版）》
CAS
北大核心
2007年第6期43-47,共5页
Journal of Jiangsu University of Science and Technology:Natural Science Edition
基金
国家自然科学基金(60572034)
江苏省自然科学基金(BK2004058)
关键词
图像检索
模糊直方图
基于区域
隶属度函数
image retrieval
fuzzy histogram
based on region
membership function
