An Overview of FHSI: Fuzzy Color Space for Image Retrieval and Modeling Aesthetic Perceptions

In this work, we provide an overview of FHSI (Fuzzy HSI), a uniform fuzzy color space for content-based image retrieval. in FHSI, colors are modeled considering the imprecision, subjectivity, context dependency and non-uniformity of color distributions. By performing color feature extraction of images, FHSI is further exploited in the application of image indexing and retrieval. This method can help to avoid time-consuming manual image tagging, providing automatic labelling. In addition, it was successfully used for quantification of the harmony and preference phenomena for the intended application of apparel coordination. We also developed a software and a library implementing the model. Experimental results prove that the proposed method results in a good retrieval efficiency and predictions of color harmony and preference.