Multipurpose watermarking is studied for both copyright protection and content authentication. In this paper, a novel multipurpose image watermarking method based on adaptive quantization of wavelet coefficients is presented. A multipurpose binary watermark tree is divided into stem, branch and leaf sub-watermarks. They are corresponding to robust, semi-fragile, and fragile watermarks, respectively. The high frequency wavelet coefficients of original image, which are divided into sub blocks with size of 3?3, are selected to carry these three sub-watermarks by using adaptive quantization index modulation. Different blocks use different quantization steps, and these quantization steps are adaptive to image characteristics as well as the function of sub-watermarks. Simulation results show that this method can get a watermarked image with perceptual invisibility. Moreover, strong robustness of stem watermark and fragility of leaf watermark can be achieved at the same time, while branch watermark has the tradeoff performance. Thus, the goal of multiple purposes is achieved effectively.