基于最小惯性轴的图像检索方法

IMAGE RETRIEVAL BASED ON AXIS OF LEAST INERTIA

提出一种基于最小惯性轴以及特征点之间结构关系的图像检索方法。方法能够利用形状轮廓和区域信息的特点提取特征点,然后根据其多值数据类型通过加权欧几里德距离计算图像间的相似度。方法对于形状的相似变换(平移、旋转和放缩)具有不变性,并更加准确地进行形状匹配。实验结果表明,方法具有较高的性能和检索效率。

In this paper,we present a method for image retrieval,which is based on the axis of least inertia and the component relationship among the feature points. This method is capable of extracting feature points in use of the features of both shape contour and region information,and then compute similarity degree between two images through Weighted Euclidean distance according to multi-valued type of the feature points. This method is invariant to image transformations （ translation,rotation and scaling） ,and the shape matching is more accurate. From the experimental results,we can see that this method has quite high performance and retrieval efficiency.