Histogram analysis for wavelet coefficients is powerful steganalysis for detecting the presence of secret information embedded in the wavelet coefficients. In order to improve the security, a wavelet-based steganography against histogram analysis is presented. First, a cover image is divided into blocks, and every block is decomposed into wavelet. Then, if the secret bit is not same as the information denoted by nonzero wavelet coefficient, the absolute value of wavelet coefficient is subtracted by 1, if the value of wavelet coefficient was 0 after embedding the secret bit, the secret bit should be embedded into next wavelet coefficient. If the sum of wavelet coefficients is large, the wavelet coefficients of next level should be embedded by secret information. Finally, the stego-image can be obtained by using the inverse wavelet transform. From the experimental results, the proposed method could effectively keep the identity of histogram for wavelet coefficients and maintain a good visual quality of stego-image.