Efficient Content-Based Image Retrieval

本研究项目的目标是设计和实现一个基于内容的图像检索系统，该系统可以(1)提供多种图像距离度量，可以单独使用或组合使用，以满足广泛的用户需求；(2)提供对图像的快速访问，即使在非常大的数据库中也是如此。这项工作的重点是开发一个通用的、可伸缩的体系结构，以支持使用用户指定的距离度量快速查询非常大的图像数据库。这包括开发独立于距离测量的算法和数据结构，以便从大型数据库中高效地检索图像。正在寻求将通用的、与距离测量无关的算法与其他可能特定于距离测量的有用技术(例如关键字检索和关系索引)相结合的方法。正在研究为用户提供多种不同品种的距离测量的问题。正在设计一种新的方法，将距离测量与一种语言相结合，用户可以用这种语言指定他们的查询，而无需详细了解基本指标。正在实现一个原型系统来测试所开发的方法，并在一个大型通用图像数据库和一个较小的受控数据库上进行了评估。这项研究的结果将是：(1)通过从搜索中消除图像数据库的巨大部分来促进图像快速检索的技术，使得基于内容的检索在非常大和不断增长的数据库上是可行的；(2)新的高级方法，通过该方法，用户可以结合距离度量来形成有意义的查询，从而基于内容的查询可以成为查询图像数据库的标准方式；以及(3)基于内容的检索的通用框架，该框架可以适应由其他研究努力开发的新的距离度量。这部作品适用于医学、艺术、摄影、娱乐和广告/营销。

The goal of this research project is to design and implement a system for content-based image retrieval that can (1) provide a large variety of image-distance measures that can be used singly or in combination to satisfy a wide range of user needs and (2) provide rapid access to images, even in an extremely large database. The focus of the work is the development of a general, scalable architecture to support fast querying of very large image databases with user-specified distance measures. This includes the development of distance-measure-independent algorithms and data structures for efficient image retrieval from large databases. Methods for merging the general, distance-measure-independent algorithms with other useful techniques that may be distance measure specific, such as keyword retrieval and relational indexing, are being pursued. The problem of providing users with multiple distance measures of many different varieties is being studied. New methods for combining distance measures and a language in which users can specify their queries without detailed knowledge of the underlying metrics are being designed. A prototype system is being implemented to test the developed methods, and evaluation is being performed on both a large general image database and a smaller controlled database. The results of this research will be: (1) techniques that facilitate rapid retrieval of images by eliminating huge portions of the image database from the search, making content-based retrieval feasible on very large and growing databases; (2) new, high-level methods by which users can combine distance measures to form meaningful queries, so that content-based queries can become a standard way to query image databases; and (3) a general framework for content-based retrieval that can accommodate new distance measures as they are developed by other research efforts. The work has application to medicine, art, photography, entertainment, and advertising/marketing.