The work aims to propose a pixel quantization steganography algorithm based on gray image, so as to improve the scale, embedding, security and robustness of gray image processing. Firstly, Hilbert curve scanning and image pixel quantization were performed on the carrier image, and the threshold region generated by the quantization information was accurately divided and marked. Then, Hilbert scrambling and D-S (data encryption security algorithm) data encryption processing were performed on the carrier image and the information to be embedded, respectively. Finally, the secret information generated after the encryption and the related auxiliary information were embedded in the ciphertext image. The experimental results showed that, the embedding rate of the proposed algorithm was increased by 10% compared with the original algorithm. The receiver could decrypt the original information and restore the complete carrier image through the key. The steganography algorithm has high information embedding rate and stronger robustness and security.