相关反馈技术在图像检索系统中的应用

Application of Relevance Feedback Techniques in Content-Based Image Retrieval

讨论了相关反馈的交互过程。分析了相关反馈中的学习问题及其特点。根据相关反馈算法所采用的检索模型把算法分为查询点移动、权系数调整、基于传统的统计学习理论和基于机器学习理论。并据此分类对近年具来有代表性的一些算法进行了分析和探讨。

This paper discusses the interactive process, and analyzes the learning problems and characteristics of relevance feedback. Based on the retrieval model adopted in the algorithm, relevance feedback algorithms are categorized into four classes, such as distance-based approach, probabilistic approach, tradition-based learning approach and machine-based learning approach. Finally, various representative algorithms are discussed according to this categorization.