ImageRetrieval Basedon FuzzyColorSemantics

Abstract-In order to improve the performance ofcontent-based image retrieval (CBIR) systems, the 'semanticgap' between the low-level visual features and the high-levelsemantic features attracts more and more research interest. Wepropose an approach to describe andto extract the fuzzy color semantics. According to humancolor perception model,we utilize the linguistic variable to describe the image color semantics, so it becomes possible to depict the image inlinguistic expression suchasmostlyred. Furthermore,weapplythe feedforward neural network to model the vagueness of human color perception and to extract the fuzzy semantic feature vector. Ourexperiments showthat the color semanticfeatures have good accordance with thehuman perception, andalso have good retrieval performance. In some extent, ourapproach shows the potential to reduce the semantic gap in CBIR. I. INTRODUCTION Content-based image retrieval systems search imagesbased on visual features, such as color, texture, shape andspatial feature. However,