A novel adaptive quantization index modulation digital image watermarking algorithm in contourlet transform domain is proposed. The original binary watermark image is preprocessed through extended scrambling, dimension reduction and pseudorandom sequence encryption. The secret watermark information bits are embedded into low-pass subband coefficients in contourlet transform domain. According to energy distribution in lowpass subband, the largest singular value of the selected lowpass subband coefficients blocks is adaptively operated by quantization index modulation for embedding watermark information. Experimental results show that the algorithm has good imperceptibility and robustness against the attacks of JPEG compression, low-pass filtering, additive noise, resampling and image cropping. Because of quantization embedding strategies, no host image is needed in watermarking extraction, which is a blind watermarking algorithm.