基于边缘方向直方图的形状检索

Shape retrieval based on edge direction histogram

近年来,随着多媒体技术和数字设备的出现,如何有效地管理和访问图像信息已成为人们亟待解决的问题。因此,一种新的图像检索技术——基于内容的图像检索技术被提出来。其中,由于图像的形状特征更符合人们的视觉感知,因此基于形状的图像检索越来越受到研究者的关注。旨在研究基于形状轮廓特征的图像检索,提出了基于边缘方向的直方图形状检索算法。通过对常用边缘检测算子的分析和比较,给出了边缘方向直方图特征提取的具体实现技术,对采用的特征匹配方法做了描述,最后通过实验的结果与分析验证了算法的性能。

With the emergence of multimedia technology and digital equipments in recent years, how to effectively manage and access image information has become an urgent problem. Therefore, a new technology, namely the content-based image retrieval technology was proposed. Because the shape features of images are more in line with the visual perception of people, the shape-based image retrieval has attracted the attention of more and more researchers. It aims to study the image retrieval based on the shape contour features, and proposed the histogram shape retrieval algorithm based on the edge direction. Through analysis and comparison of common edge retrieval operators, this paper proposed the specific technology of extracting the features of edge direction histogram, described the feature matching method adopted, and finally verified the performance of the algorithm through the results of experiments and analysis.