基于内容的图像检索技术

Technique of content-based image retrieval

目前各行业对图像的使用越来越广泛,如何有效、快速地从大规模图像数据库中检索出需要的图像,是目前一个相当重要而又富有挑战性的研究课题。但传统的图像检索技术是基于文本的检索技术,这种方法虽然简单易行,但存在一些致命的缺点,严重影响了对图像信息的有效使用。为了克服传统方法的缺点,提出了基于内容的图像检索技术,该技术能够全面客观地提取图像内容,能有效地获取所需的视觉信息,能使图像数据库中的信息得到有效的管理。

Recently, images are widely used in many fields. How to search for desired images from large-scale image databases rapidly and efficiently is an important and challenging research topic. Traditional image retrieval is based on text, this technique is easy but it has some deadly shortcomings, which affect the efficient use of images＇ information severely. The paper presents content-based image retrieval, which can distill image content comprehensively and impersonally. This technique also can get required vision information and manage the information of image databases efficiently.