完整的基于DWT的数字图像信息隐藏算法代码，带有高斯噪声攻击，可以运行，载体图像和水印图像经过灰度处理，有提取和嵌入水印算法，用python实现

以下是基于DWT的数字图像信息隐藏算法的Python实现代码，包含高斯噪声攻击、提取和嵌入水印算法：

import numpy as np
import cv2
from scipy.fftpack import dct, idct
from scipy.stats import norm


def embed_watermark(img_path, wm_path, alpha):
    # 读取载体图像和水印图像
    img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)
    wm = cv2.imread(wm_path, cv2.IMREAD_GRAYSCALE)

    # 对载体图像和水印图像进行离散小波变换
    img_w = pywt.wavedec2(img, 'haar', level=1)
    wm_w = pywt.wavedec2(wm, 'haar', level=1)

    # 将水印图像进行0-1归一化
    wm_norm = wm_w[0] / 255

    # 对LL子带进行DCT变换
    img_LL = img_w[0]
    img_LL_dct = dct(img_LL, axis=0)
    img_LL_dct = dct(img_LL_dct, axis=1)

    # 对DCT系数进行量化
    Q = np.array([[16, 11, 10, 16, 24, 40, 51, 61],
                  [12, 12, 14, 19, 26, 58, 60, 55],
                  [14, 13, 16, 24, 40, 57, 69, 56],
                  [14, 17, 22, 29, 51, 87, 80, 62],
                  [18, 22, 37, 56, 68, 109, 103, 77],
                  [24, 35, 55, 64, 81, 104, 113, 92],
                  [49, 64, 78, 87, 103, 121, 120, 101],
                  [72, 92, 95, 98, 112, 100, 103, 99]])
    img_LL_dct = np.round(img_LL_dct / (alpha * Q)) * (alpha * Q)

    # 将归一化后的水印嵌入到DCT系数的最低位
    wm_embed = np.zeros_like(img_LL_dct)
    wm_embed[:, :, 0] = img_LL_dct[:, :, 0]
    wm_embed[:, :, 1] = img_LL_dct[:, :, 1]
    wm_embed[:, :, 2] = img_LL_dct[:, :, 2]
    wm_embed[:, :, 3] = img_LL_dct[:, :, 3]
    wm_embed[:, :, 4] = img_LL_dct[:, :, 4]
    wm_embed[:, :, 5] = img_LL_dct[:, :, 5]
    wm_embed[:, :, 6] = img_LL_dct[:, :, 6]
    wm_embed[:, :, 7] = img_LL_dct[:, :, 7]
    wm_embed[:, :, 0] += 2 * wm_norm * alpha
    wm_embed[:, :, 1] += 2 * wm_norm * alpha
    wm_embed[:, :, 2] += 2 * wm_norm * alpha
    wm_embed[:, :, 3] += 2 * wm_norm * alpha
    wm_embed[:, :, 4] += 2 * wm_norm * alpha
    wm_embed[:, :, 5] += 2 * wm_norm * alpha
    wm_embed[:, :, 6] += 2 * wm_norm * alpha
    wm_embed[:, :, 7] += 2 * wm_norm * alpha

    # 对嵌入水印后的DCT系数进行反变换
    img_LL_dct_embed = idct(wm_embed, axis=0)
    img_LL_dct_embed = idct(img_LL_dct_embed, axis=1)

    # 对反变换得到的图像进行量化逆操作
    img_LL_dct_embed = np.round(img_LL_dct_embed / Q) * Q

    # 对量化逆操作后的图像进行反DCT变换
    img_LL_embed = idct(img_LL_dct_embed, axis=0)
    img_LL_embed = idct(img_LL_embed, axis=1)

    # 将嵌入水印后的LL子带与其他子带组合成新的小波系数
    img_w_embed = list(img_w)
    img_w_embed[0] = img_LL_embed
    img_embed = pywt.waverec2(img_w_embed, 'haar')

    # 对嵌入水印后的图像进行高斯噪声攻击
    img_embed = np.float32(img_embed)
    img_embed_noise = cv2.randn(img_embed, 0, 10)
    img_embed_noise = np.clip(img_embed_noise, 0, 255)
    img_embed_noise = np.uint8(img_embed_noise)

    # 返回嵌入水印后的图像和添加高斯噪声的图像
    return img_embed, img_embed_noise


def extract_watermark(img_path, alpha):
    # 读取载体图像
    img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)

    # 对载体图像进行离散小波变换
    img_w = pywt.wavedec2(img, 'haar', level=1)

    # 对LL子带进行DCT变换
    img_LL = img_w[0]
    img_LL_dct = dct(img_LL, axis=0)
    img_LL_dct = dct(img_LL_dct, axis=1)

    # 对DCT系数进行量化
    Q = np.array([[16, 11, 10, 16, 24, 40, 51, 61],
                  [12, 12, 14, 19, 26, 58, 60, 55],
                  [14, 13, 16, 24, 40, 57, 69, 56],
                  [14, 17, 22, 29, 51, 87, 80, 62],
                  [18, 22, 37, 56, 68, 109, 103, 77],
                  [24, 35, 55, 64, 81, 104, 113, 92],
                  [49, 64, 78, 87, 103, 121, 120, 101],
                  [72, 92, 95, 98, 112, 100, 103, 99]])
    img_LL_dct_q = np.round(img_LL_dct / (alpha * Q)) * (alpha * Q)

    # 提取嵌入的水印
    wm_extract_norm = (img_LL_dct_q[:, :, 0] % (2 * alpha)) / (2 * alpha)

    # 将提取的水印还原到0-255范围内
    wm_extract = np.uint8(wm_extract_norm * 255)

    # 返回提取的水印
    return wm_extract

注意：在使用该代码时，需要先安装依赖库`numpy`、`scipy`和`opencv-python`。另外，该代码中的`alpha`参数用于控制嵌入水印的强度，需要根据实际情况进行调整。