基于颜色特征分布及相关反馈的图像检索方法

在基于内容的图像检索中,大部分工作主要集中在提高检索系统的独立性和全自动性方面,由于存在着低层语义特征与高级概念间的差距、视觉感知的主观性等因素影响,使传统检索方法受到了限制,在巳提出的基于颜色空间分布检索方法基础上,引入相关反馈机制,弥补了原有方法的不足,进一步体现了系统的交互性,提高了图像检索效果。

Content-based Image Retrieval Based on Color Feature Distribution and Relevance Feedback

Most of the early literatures emphasize "fully automated system" in CBIR. In fact, in the computer centric approach, the "best" features and representations and their corresponding weights are fixed, which can't effectively model low level features and user's perception subjectivity. Recent research focus has moved to an interactive mechanism that involves a human as part of the retrieval process. In this paper, a relevance feedback mechanism is cited in content based image retrieval based on color spatial distribution. The experimental results show that the proposed approach greatly improves the image retrieval's effect.