Encrypted Image Retrieval with Scalable Color Histograms

Abstract

In order to address the problem of image retrieval from an encrypted database, this paper proposes an encrypted image retrieval scheme based on scalable color histogram. The plaintext images are firstly encrypted by permuting pixels’ position to preserve image content. For any encrypted image, a scalable color histogram is obtained by dividing the pixel value to different gray level, although the server does not know the image content. By utilizing the scalable color histogram, the integrated distance is calculated and compared between the encrypted query image and every encrypted image in database. Experimental results show that proposed retrieval scheme can achieve better retrieval performance than plaintext retrieval scheme, and provide effective protection for users’ privacy.