Image Robust Hashing Based on DCT Sign

Image robust hashing is related to cryptographic hash functions. In contrast to cryptographic hash functions this robust digest is sensitive only to perceptual change. Minor changes, which are not affecting the perception, do not result in a different hash. Image robust hashing is used in content-based retrieval, monitoring, and filtering. In this paper an image robust hashing method based on sign bit from the domain of Discrete Cosine Transform (DCT) is presented. DCT is widely used in image processing and video processing, e.g. for compression or digital watermarking. From an image with reduced dimensions we apply 2-D DCT to derive an initial feature vector. The sign bit of this feature vector is extracted to form an intermediate hash. The intermediate hash can be incorporated into some security mechanism to derive a final hash. The advantages of the sign signal in DCT domain are verified in experiments evaluating robustness (e.g. against operations like lossy compression, scaling and cropping) and discriminability. The results show a better performance of the proposed method in this paper than an existing method.