Authentication watermarking has been the most effective method to ensure trustworthiness of multimedia data, and it has been developed and become an exciting research field. Semi-fragile watermarking technique is an important branch of authentication watermarking. Semi-fragile watermarking is robust for common signal processing operations, but fragile for intent changing manipulations. It can distinguish between general image processing and tampered image.The hash function has the fixed long output value, it is suitable for authentication as authentication value. As a new digital multimedia authentication technology, hash based semi-fragile watermarking has been greatly developed for these years. The algorithms of authentication in this paper are hash based semi-fragile watermarking.In this thesis, the authentication watermarking techniques of images are discussed, and the main content and contributions of this dissertation are: First, in this paper, perceptual hash based image watermarking algorithm with hash function and human visual masking is proposed to achieve image-content based authentication. It uses end-stopped wavelet to exact feature points in images, and contains feature points as hash values, finally uses hash values to authenticate. To enhance robustness of geometric attacks, radon transformation of edge image is used to detect and invert geometric transformations, and a perceptual hash feature detector is proposed to extract significant geometry preserving feature points, finally compare features between two images by developing a modified Hausdorff distance measure.The experimental results show that the algorithm is robust for JPEG compression and geometric transformation, and can detect the tampered exact position of location. Second, a dual encryted watermarking algorithm which based on the hash algorithm is proposed. It is semi-fragile watermarking which is based on hash algorithm, in this paper, in combination with the DCT and DWT transform features, carries on the dual encryptions using the magic transformation and the Arnold transform, forms the final key and the hash value pushes the image to carry on the authentication. The experimental results show that the algorithm enhances the performance of JPEG anti-compression, and can detect the exact tampered position and location. In addition, the dual keys produced from the algorithm results in better security for the algorithm. |