摘要
基于内容的图像检索是当前国内外的一个研究热点。形状是物体的一个重要特征,是刻画物体结构、形态、轮廓的特征量,也是比较难描述的图像特征之一。由于物体形状的自动获取涉及图像中感兴趣目标的分割,所以基于形状检索的图像检索是一项有难度的研究工作。基于形状的检索一般适用于较容易识别的物体。在本文中采用了面积、形状矩等局部或全局特征来表示一幅图像。实验系统表明,本文采用的特征能有效地实现汽车等目标的检索。
Content-based image retrieval is a research focus at home and abroad.Shape is an important feature of the object in an image.It describes the object＇s features such as structure,pattern,contour etc.It is one of the most difficult features to be described.Because the automatic acquisition of objects＇ shape is related to segmentation of objects of interest in image,shape-based retrieval is a difficult research.Shape-based retrieval is generally applicable to the objects which can be easily identified.In this paper,area,shape moments and other local or global features are used to represent an image.Experimental results show these features can efficiently realize the retrieval of objects such as cars.
出处
《国外电子测量技术》
2011年第10期22-24,36,共4页
Foreign Electronic Measurement Technology
基金
江苏省"六大人才高峰"第七批高层次人才项目资助
关键词
基于内容的图像检索
特征提取
相似性度量
CBIR （content-based image retrieval）
feature extraction
similarity measurement
