An Enhanced Perceptual Image Hashing Scheme based on Clustering
Yang Daniel Ou, Chul Sur, Kyung-Hyune Rhee · 한국멀티미디어학회 학술발표논문집 · 2009
In this paper, we propose a perceptual image hashing scheme by using image histogram and K-means clustering algorithm. Our work is motivated by previous histogram-based image hashing scheme which achieves a good robustness to perceptually insignificant attack but the image structure is lost during hash generation. We employ the K-means algorithm to cluster the image into two or more segments, and calculate the histogram in each segment, separately. The clustering algorithm can capture local image information and make our scheme be more fragile to perceptual significant attacks. The experimental results confirm that our proposed scheme not only preserves the robustness of original one, but also enhances the fragility to malicious attacks.