Content-based image retrieval base on relevance feedback
Haixin Wen, Yinwei Zhan · AIP conference proceedings · 2017
In order to reduce the “semantic gap” between the underlying features and the high-level semantics in the image retrieval, the feature re-weighted correlation feedback mechanism is applied. The basic idea is to reduce the weight of the irrelevant feature dimension in the feature space and increase the weight of the relevant feature dimension. The improved method of feature weight adjustment proposed in this paper is compared with another method. The accuracy of the algorithm is proved by the experiment as the criterion of the evaluation effect.