SAM_PRO_super_resolution_visualization.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7855ca3c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import hdf5storage\n",
    "import cv2\n",
    "import matlab\n",
    "import matlab.engine\n",
    "from PnP_restoration.utils.utils_restoration import single2uint,crop_center, matlab_style_gauss2D, imread_uint, imsave\n",
    "from scipy import ndimage\n",
    "from PIL import Image\n",
    "\n",
    "def CalMATLAB(IRFolder,GTFolder):\n",
    "    eng = matlab.engine.start_matlab()\n",
    "    eng.addpath(eng.genpath(eng.fullfile(os.getcwd(),'Metrics')))\n",
    "    res = eng.evaluate_PSNR(IRFolder,GTFolder)\n",
    "    res=np.array(res)\n",
    "    return res\n",
    "\n",
    "def CalMATLAB2(IRFolder,one_gt_name):\n",
    "    eng = matlab.engine.start_matlab()\n",
    "    eng.addpath(eng.genpath(eng.fullfile(os.getcwd(),'Metrics')))\n",
    "    res = eng.evaluate_PSNR_comparison(IRFolder,one_gt_name)\n",
    "    res=np.array(res)\n",
    "    return res\n",
    "\n",
    "def MATLAB_imresize(imgname, Sf=3):\n",
    "    eng = matlab.engine.start_matlab()\n",
    "    x = eng.load(imgname)['data']\n",
    "#     eng.addpath(eng.genpath(eng.fullfile(os.getcwd(),'Metrics')))\n",
    "    res = eng.imresize(x, Sf)\n",
    "    res=np.array(res)\n",
    "    return res\n",
    "\n",
    "def MATLAB_imresize2(imgname, Sf=3):\n",
    "    eng = matlab.engine.start_matlab()\n",
    "    x = eng.imread(imgname)\n",
    "#     eng.addpath(eng.genpath(eng.fullfile(os.getcwd(),'Metrics')))\n",
    "    res = eng.imresize(x, Sf)\n",
    "    res=np.array(res)\n",
    "    return res\n",
    "\n",
    "\n",
    "def MATLAB_degradation(imgname, Sf=3):\n",
    "    eng = matlab.engine.start_matlab()\n",
    "    x = eng.imread(imgname)\n",
    "    x = matlab.double(x)\n",
    "#     eng.addpath(eng.genpath(eng.fullfile(os.getcwd(),'Metrics')))\n",
    "    res = eng.imresize(x, Sf)\n",
    "    res=np.array(res)\n",
    "    return res\n",
    "\n",
    "\n",
    "def mkdir(path):\n",
    "    if not os.path.exists(path):\n",
    "        os.makedirs(path)\n",
    "\n",
    "def norm_proj(x):\n",
    "    x = (x-np.min(x))/(np.max(x)-np.min(x))\n",
    "    return x\n",
    "\n",
    "def imsave(img, img_path):\n",
    "    if img.ndim == 3:\n",
    "        img = img[:, :, [2, 1, 0]]\n",
    "    cv2.imwrite(img_path, img)\n",
    "    \n",
    "def numpy_degradation(x, k, sf=3):\n",
    "    ''' blur + downsampling\n",
    "    Args:\n",
    "        x: HxWxC image, [0, 1]/[0, 255]\n",
    "        k: hxw, double, positive\n",
    "        sf: down-scale factor\n",
    "    Return:\n",
    "        downsampled LR image\n",
    "    '''\n",
    "    x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap')\n",
    "    # x = filters.correlate(x, np.expand_dims(np.flip(k), axis=2))\n",
    "    st = 0\n",
    "    return x[st::sf, st::sf, ...]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c028dab6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_46797/408656523.py:74: DeprecationWarning: Please use `convolve` from the `scipy.ndimage` namespace, the `scipy.ndimage.filters` namespace is deprecated.\n",
      "  x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap')\n"
     ]
    }
   ],
   "source": [
    "kernel_path = os.path.join('PnP_restoration/kernels', 'Levin09.mat')\n",
    "kernels = hdf5storage.loadmat(kernel_path)['kernels']\n",
    "name = 'parrots'###bike, butterfly,flower, girl, hat,parrots\n",
    "k_index = 8\n",
    "Sf = 3\n",
    "task = 'super-resolution'\n",
    "img_type = 'RGB'\n",
    "noise_level = 5#2**0.5 #2**0.5\n",
    "algorithm = 'SAM_PROv1'#################################SAM_PROv1, RED, RED_PRO, PnP_FBS##################################\n",
    "k = matlab_style_gauss2D(shape=(7,7),sigma=1.6)\n",
    "if img_type == 'RGB':\n",
    "    input_im_uint = imread_uint('test_images/'+name+'.tif',n_channels=3)\n",
    "#     input_im_uint = imread_uint('datasets/set12/05.png', n_channels=3)\n",
    "else:\n",
    "    input_im_uint = imread_uint('test_images/'+name+'.tif',n_channels=1)\n",
    "#     input_im_uint = imread_uint('datasets/set12/05.png', n_channels=1)\n",
    "input_im = np.float32(input_im_uint / 255)\n",
    "            # Degrade image\n",
    "if Sf>1:#super-resolution\n",
    "    blur_im = numpy_degradation(input_im, k, sf=Sf)\n",
    "else:# debluring\n",
    "    blur_im = ndimage.filters.convolve(input_im, np.expand_dims(k, axis=2), mode='wrap')\n",
    "np.random.seed(seed=0)\n",
    "noise = np.random.normal(0, noise_level / 255., blur_im.shape)\n",
    "blur_im += noise\n",
    "init_im = blur_im"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d3fba4c9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(256, 256, 3) (86, 86, 3)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x1600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(6,16))\n",
    "if img_type == 'RGB':\n",
    "    ax[0].imshow(input_im_uint)\n",
    "    # ax[0].set_title('Input image')\n",
    "    ax[0].set_axis_off()\n",
    "    ax[1].imshow(np.uint8(norm_proj(init_im)*255))\n",
    "    # ax[1].set_title('Dice loss')\n",
    "    ax[1].set_axis_off()\n",
    "else:\n",
    "    ax[0].imshow(input_im,cmap='Greys_r')\n",
    "    # ax[0].set_title('Input image')\n",
    "    ax[0].set_axis_off()\n",
    "    ax[1].imshow(init_im, cmap='Greys_r')\n",
    "    # ax[1].set_title('Dice loss')\n",
    "    ax[1].set_axis_off()\n",
    "print(input_im_uint.shape,init_im.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6383cfb1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "from PnP_restoration.utils.utils_restoration import rgb2y, psnr, array2tensor, tensor2array\n",
    "from PnP_restoration.utils import utils_sr\n",
    "from utils.utils import load_model\n",
    "from utils.utils import psnr\n",
    "from utils.config import analyze_parse\n",
    "import math\n",
    "\n",
    "def initialize_prox(img, degradation_mode, degradation, sf, device):\n",
    "#     degradation_mode = degradation\n",
    "    if degradation_mode == 'deblurring':\n",
    "        k = degradation\n",
    "        k_tensor = array2tensor(np.expand_dims(k, 2)).double().to(device)\n",
    "        FB, FBC, F2B, FBFy = utils_sr.pre_calculate_prox2(img, k_tensor, sf)\n",
    "        return FB, FBC, F2B, FBFy, k_tensor\n",
    "    elif degradation_mode == 'SR':\n",
    "        k = degradation\n",
    "        k_tensor = array2tensor(np.expand_dims(k, 2)).double().to(device)\n",
    "        FB, FBC, F2B, FBFy = utils_sr.pre_calculate_prox2(img,k_tensor, sf)\n",
    "        return FB, FBC, F2B, FBFy, k_tensor\n",
    "    elif degradation_mode == 'inpainting':\n",
    "        M = array2tensor(degradation).double().to(device)\n",
    "        My = M*img\n",
    "        return My\n",
    "    else:\n",
    "        print('degradation mode not treated')\n",
    "\n",
    "def calulate_data_term(k_tensor,degradation_mode, sf,y,img):\n",
    "        '''\n",
    "        Calculation of the data term value f(y)\n",
    "        :param y: Point where to evaluate F\n",
    "        :param img: Degraded image\n",
    "        :return: f(y)\n",
    "        '''\n",
    "#         k_tensor = array2tensor(np.expand_dims(k, 2)).double().to(device)\n",
    "        if degradation_mode == 'deblurring':\n",
    "            deg_y = utils_sr.imfilter(y.double(), k_tensor[0].double().flip(1).flip(2).expand(3, -1, -1, -1))\n",
    "            f = 0.5 * torch.norm(img - deg_y, p=2) ** 2\n",
    "        elif degradation_mode == 'SR':\n",
    "            deg_y = utils_sr.imfilter(y.double(), k_tensor[0].double().flip(1).flip(2).expand(3, -1, -1, -1))\n",
    "            deg_y = deg_y[..., 0::sf, 0::sf]\n",
    "            f = 0.5 * torch.norm(img - deg_y, p=2) ** 2\n",
    "#         elif degradation_mode == 'inpainting':\n",
    "#             deg_y = M * y.double()\n",
    "#             f = 0.5 * torch.norm(img - deg_y, p=2) ** 2\n",
    "        else:\n",
    "            print('degradation not implemented')\n",
    "        return f\n",
    "        \n",
    "def rgb2ycbcr(img, only_y=True):\n",
    "    '''same as matlab rgb2ycbcr\n",
    "    only_y: only return Y channel\n",
    "    Input:\n",
    "        uint8, [0, 255]\n",
    "        float, [0, 1]\n",
    "    '''\n",
    "    in_img_type = img.dtype\n",
    "    img.astype(np.float32)\n",
    "    if in_img_type != np.uint8:\n",
    "        img *= 255.\n",
    "    # convert\n",
    "    if only_y:\n",
    "        rlt = np.dot(img, [65.481, 128.553, 24.966]) / 255.0 + 16.0\n",
    "    else:\n",
    "        rlt = np.matmul(img, [[65.481, -37.797, 112.0], [128.553, -74.203, -93.786],\n",
    "                              [24.966, 112.0, -18.214]]) / 255.0 + [16, 128, 128]\n",
    "    if in_img_type == np.uint8:\n",
    "        rlt = rlt.round()\n",
    "    else:\n",
    "        rlt /= 255.\n",
    "    return rlt.astype(in_img_type)\n",
    "\n",
    "\n",
    "def ycbcr2rgb(img):\n",
    "    '''same as matlab ycbcr2rgb\n",
    "    Input:\n",
    "        uint8, [0, 255]\n",
    "        float, [0, 1]\n",
    "    '''\n",
    "    in_img_type = img.dtype\n",
    "    img.astype(np.float32)\n",
    "    if in_img_type != np.uint8:\n",
    "        img *= 255.\n",
    "    # convert\n",
    "    rlt = np.matmul(img, [[0.00456621, 0.00456621, 0.00456621], [0, -0.00153632, 0.00791071],\n",
    "                          [0.00625893, -0.00318811, 0]]) * 255.0 + [-222.921, 135.576, -276.836]\n",
    "    rlt = np.clip(rlt, 0, 255)\n",
    "    if in_img_type == np.uint8:\n",
    "        rlt = rlt.round()\n",
    "    else:\n",
    "        rlt /= 255.\n",
    "    return rlt.astype(in_img_type)\n",
    "\n",
    "def single2uint(img):\n",
    "    return np.uint8(img*255.)\n",
    "\n",
    "def tensor2uint(img):\n",
    "    img = img.data.squeeze().float().clamp_(0, 1).cpu().numpy()\n",
    "    if img.ndim == 3:\n",
    "        img = np.transpose(img, (1, 2, 0))\n",
    "    return np.uint8((img*255.0).round())\n",
    "\n",
    "def tensor2uint2(img):\n",
    "    img = img.data.squeeze().float().cpu().numpy()\n",
    "    img = norm_proj(img)\n",
    "    if img.ndim == 3:\n",
    "        img = np.transpose(img, (1, 2, 0))\n",
    "    return np.uint8((img*255.0).round())\n",
    "\n",
    "def tensor2float(img):\n",
    "    img = img.data.squeeze().float().clamp_(0, 1).cpu().numpy()\n",
    "    if img.ndim == 3:\n",
    "        img = np.transpose(img, (1, 2, 0))\n",
    "    return img\n",
    "\n",
    "def tensor2float2(img):\n",
    "    img = img.data.squeeze().float().cpu().numpy()\n",
    "    img = norm_proj(img)\n",
    "    if img.ndim == 3:\n",
    "        img = np.transpose(img, (1, 2, 0))\n",
    "    return img\n",
    "    \n",
    "def calculate_psnr(img1, img2, border=0):\n",
    "    # img1 and img2 have range [0, 255]\n",
    "    if not img1.shape == img2.shape:\n",
    "        raise ValueError('Input images must have the same dimensions.')\n",
    "    h, w = img1.shape[:2]\n",
    "    img1 = img1[border:h-border, border:w-border]\n",
    "    img2 = img2[border:h-border, border:w-border]\n",
    "\n",
    "    img1 = img1.astype(np.float64)\n",
    "    img2 = img2.astype(np.float64)\n",
    "    mse = np.mean((img1 - img2)**2)\n",
    "    if mse == 0:\n",
    "        return float('inf')\n",
    "    return 20 * np.log10(255.0 /np.sqrt(mse))\n",
    "\n",
    "def calculate_grad(img, degradation_mode, FB, FBC,FBFy,sf=1):\n",
    "    if degradation_mode == 'deblurring':\n",
    "        grad = utils_sr.grad_solution2(img.double(), FB, FBC, FBFy, 1)\n",
    "    if degradation_mode == 'SR' :\n",
    "        grad = utils_sr.grad_solution2(img.double(), FB, FBC, FBFy, sf)\n",
    "    return grad\n",
    "\n",
    "def load_model(model_type, sigma):\n",
    "    path = \"Pretrained_models/\" + model_type + \"_noise\" + str(sigma) + \".pth\"\n",
    "    if model_type == \"DnCNN\":\n",
    "        from model.models import DnCNN\n",
    "        net = DnCNN(channels=1, num_of_layers=17)\n",
    "        model = nn.DataParallel(net).cuda()\n",
    "    elif model_type == \"SimpleCNN\":\n",
    "        from model.SimpleCNN_models import DnCNN\n",
    "        model = DnCNN(1, num_of_layers = 4, lip = 0.0, no_bn = True).cuda()\n",
    "    elif model_type == \"RealSN_DnCNN\":\n",
    "        from model.realSN_models import DnCNN\n",
    "        net = DnCNN(channels=1, num_of_layers=17)\n",
    "        model = nn.DataParallel(net).cuda()\n",
    "    elif model_type == \"RealSN_SimpleCNN\":\n",
    "        from model.SimpleCNN_models import DnCNN\n",
    "        model = DnCNN(1, num_of_layers = 4, lip = 1.0, no_bn = True).cuda()\n",
    "    else:\n",
    "        from model.realSN_models import DnCNN\n",
    "        net = DnCNN(channels=1, num_of_layers=17)\n",
    "        model = nn.DataParallel(net).cuda()\n",
    "\n",
    "    model.load_state_dict(torch.load(path))\n",
    "    model.eval()\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a7ab8822",
   "metadata": {},
   "outputs": [],
   "source": [
    "def jiasu_hsd_proj_obj_fun2(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    sigma = opt['sigma']\n",
    "    mu_0 = opt['mu_0']\n",
    "    Sf = opt['sf']\n",
    "#     y = x0.clone()\n",
    "#     print(Sf)\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu_k = mu_0*(i+1)**(-1.0)\n",
    "        mu = mu_k if mu_k<1 else 1\n",
    "        ####### T_beta\n",
    "#         grad1 =  calculate_grad(v_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        grad1 =  calculate_grad(x, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        f_est = x - alpha*grad1/sigma\n",
    "#         f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "        #         xtilde = max(min(xtl))\n",
    "        #         index = 0<=xtilde& xtilde<=1\n",
    "        #         xtilde = xtilde\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z = xtilde - r\n",
    "#         z = z.clamp_(0, 1)\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "#         r = f(xtilde.float())\n",
    "#         z= xtilde - r\n",
    "#         z = (1-beta)*f_est+beta*z\n",
    "        grad =  calculate_grad(z, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        v_est = z - mu*grad/sigma\n",
    "        w = v_est\n",
    "        \n",
    "#         print(torch.norm(w-x0, p=2))\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "#         r0 = residual.float().cpu().numpy()\n",
    "#         print(residual)\n",
    "        x0 = w\n",
    "        ########\n",
    "        obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def SAM_PRO_v2(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    sigma = opt['sigma']\n",
    "    mu_0 = opt['mu_0']\n",
    "    Sf = opt['sf']\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu_k = mu_0*(i+1)**(-1.0)\n",
    "        mu = mu_k if mu_k<1 else 1\n",
    "        f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z= xtilde - r\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "        z = (1-beta)*f_est+beta*z\n",
    "        grad =  calculate_grad(z, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        v_est = z - mu*grad/sigma\n",
    "        w = v_est\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "        x0 = w\n",
    "        obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def RED(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    lambdaa = opt['lambda']\n",
    "    sigma = opt['sigma']\n",
    "    input_sigma = opt['input_sigma']\n",
    "    Sf = opt['sf']\n",
    "#     y = x0.clone()\n",
    "#     print(Sf)\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "#     mu = 2/(1/(input_sigma**2) + lambdaa)\n",
    "    mu = 0.1####### 0.1 for parrots(sacle =3, super-resolution); 1 for butterfly(sigma=8, uniform PSF)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        ####### T_beta\n",
    "#         grad1 =  calculate_grad(v_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        grad1 =  calculate_grad(x, degradation_mode, FB, FBC, FBFy, sf=Sf)/sigma\n",
    "        f_est = x \n",
    "#         f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "        #         xtilde = max(min(xtl))\n",
    "        #         index = 0<=xtilde& xtilde<=1\n",
    "        #         xtilde = xtilde\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z = xtilde - r\n",
    "#         z = z.clamp_(0, 1)\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "        grad2 =  (x - z)\n",
    "        v_est = x - mu*(grad1/sigma+lambdaa*grad2)\n",
    "        w = v_est\n",
    "        \n",
    "#         print(torch.norm(w-x0, p=2))\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "#         r0 = residual.float().cpu().numpy()\n",
    "#         print(residual)\n",
    "        x0 = w\n",
    "        ########\n",
    "#         obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "        x_square = torch.norm(x0, p=2)**2\n",
    "        obj_fun[i] = 1/sigma*calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()/x_square.float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def REDPRO(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    lambdaa = opt['lambda']\n",
    "    sigma = opt['sigma']\n",
    "    input_sigma = opt['input_sigma']\n",
    "    Sf = opt['sf']\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu =2*(i+1)**(-0.1)################# 2 for parrots(sacle =3, super-resolution); 4 for butterfly(sigma=8, uniform PSF)\n",
    "        ####10 for hat (sigma=8, gaussian PSF)\n",
    "        ####### T_beta\n",
    "#         grad1 =  calculate_grad(v_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        grad1 =  calculate_grad(x, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        f_est = x - mu*grad1/sigma\n",
    "#         f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "        #         xtilde = max(min(xtl))\n",
    "        #         index = 0<=xtilde& xtilde<=1\n",
    "        #         xtilde = xtilde\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z = xtilde - r\n",
    "#         z = z.clamp_(0, 1)\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "        v_est = beta*z+(1-beta)*f_est\n",
    "        w = v_est\n",
    "        \n",
    "#         print(torch.norm(w-x0, p=2))\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "#         r0 = residual.float().cpu().numpy()\n",
    "#         print(residual)\n",
    "        x0 = w\n",
    "        ########\n",
    "#         obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "        x_square = torch.norm(x0, p=2)**2\n",
    "        obj_fun[i] = 1/sigma*calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()/x_square.float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def PnP_FBS(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    lambdaa = opt['lambda']\n",
    "    sigma = opt['sigma']\n",
    "    input_sigma = opt['input_sigma']\n",
    "    Sf = opt['sf']\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    mu0 = 4\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu = mu0################# 2 for parrots(sacle =3, super-resolution); 4 for butterfly(sigma=8, uniform PSF)\n",
    "        ####10 for hat (sigma=8, gaussian PSF)\n",
    "        ####### T_beta\n",
    "#         grad1 =  calculate_grad(v_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        grad1 =  calculate_grad(x, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        f_est = x - mu*grad1/sigma\n",
    "#         f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "        #         xtilde = max(min(xtl))\n",
    "        #         index = 0<=xtilde& xtilde<=1\n",
    "        #         xtilde = xtilde\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z = xtilde - r\n",
    "#         z = z.clamp_(0, 1)\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "        v_est = beta*z+(1-beta)*x\n",
    "        w = v_est\n",
    "        \n",
    "#         print(torch.norm(w-x0, p=2))\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "#         r0 = residual.float().cpu().numpy()\n",
    "#         print(residual)\n",
    "        x0 = w\n",
    "        ########\n",
    "#         obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "        x_square = torch.norm(x0, p=2)**2\n",
    "        obj_fun[i] = 1/sigma*calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()/x_square.float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def jiasu_hsd_proj_obj_fun_mat(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    sigma = opt['sigma']\n",
    "    mu_0 = opt['mu_0']\n",
    "    Sf = opt['sf']\n",
    "#     y = x0.clone()\n",
    "#     print(Sf)\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu_k = mu_0*(i+1)**(-1.0);\n",
    "        mu = mu_k if mu_k<1 else 1\n",
    "        ####### T_beta\n",
    "#         grad1 =  calculate_grad(v_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "#         grad1 =  calculate_grad(x, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "#         f_est = x - alpha*grad1/sigma\n",
    "        f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "        #         xtilde = max(min(xtl))\n",
    "        #         index = 0<=xtilde& xtilde<=1\n",
    "        #         xtilde = xtilde\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z= xtilde - r\n",
    "#         z = z.clamp_(0, 1)\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "#         r = f(xtilde.float())\n",
    "#         z= xtilde - r\n",
    "#         if i>500:\n",
    "#             beta = 0.01\n",
    "#         else:\n",
    "#             beta = 0.01\n",
    "        z = (1-beta)*f_est+beta*z\n",
    "        grad =  calculate_grad(z, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        v_est = z - mu*grad/sigma\n",
    "        w = v_est\n",
    "        \n",
    "#         print(torch.norm(w-x0, p=2))\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "#         r0 = residual.float().cpu().numpy()\n",
    "#         print(residual)\n",
    "        x0 = w\n",
    "        tensor_to_mat(str(i),name, x0, setting='RealSN_SimpleCNN')\n",
    "        ########\n",
    "        obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def jiasu_hsd_proj_obj_fun_beta_mat(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    sigma = opt['sigma']\n",
    "    mu_0 = opt['mu_0']\n",
    "    Sf = opt['sf']\n",
    "#     y = x0.clone()\n",
    "#     print(Sf)\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu_k = mu_0*(i+1)**(-1.0);\n",
    "        mu = mu_k if mu_k<1 else 1\n",
    "        ####### T_beta\n",
    "#         grad1 =  calculate_grad(v_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "#         grad1 =  calculate_grad(x, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "#         f_est = x - alpha*grad1/sigma\n",
    "        f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "        #         xtilde = max(min(xtl))\n",
    "        #         index = 0<=xtilde& xtilde<=1\n",
    "        #         xtilde = xtilde\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z= xtilde - r\n",
    "#         z = z.clamp_(0, 1)\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "#         xtilde = f_est.clamp_(0, 1)\n",
    "#         r = f(xtilde.float())\n",
    "#         z= xtilde - r\n",
    "#         if i>500:\n",
    "#             beta = 0.01\n",
    "#         else:\n",
    "#             beta = 0.01\n",
    "        z = (1-beta)*f_est+beta*z\n",
    "        grad =  calculate_grad(z, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        v_est = z - mu*grad/sigma\n",
    "        w = v_est\n",
    "        \n",
    "#         print(torch.norm(w-x0, p=2))\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "#         r0 = residual.float().cpu().numpy()\n",
    "#         print(residual)\n",
    "        x0 = w\n",
    "        tensor_to_mat(str(i),name, x0, setting=str(beta))\n",
    "        ########\n",
    "        obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "\n",
    "def SAM_PRO_v1(x0,y, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt):\n",
    "    K = opt['K']\n",
    "    alpha = opt['alpha']\n",
    "    beta = opt['beta']\n",
    "    sigma = opt['sigma']\n",
    "    mu_0 = opt['mu_0']\n",
    "    Sf = opt['sf']\n",
    "    obj_fun = np.zeros(K)\n",
    "    residual = torch.zeros(K)\n",
    "    for i in range(K):\n",
    "        x = x0\n",
    "        mu_k = mu_0*(i+1)**(-1.0)\n",
    "        mu = mu_k if mu_k<1 else 1\n",
    "        f_est = x\n",
    "        mintmp = torch.min(f_est)\n",
    "        maxtmp = torch.max(f_est)\n",
    "        xtilde = (f_est - mintmp) / (maxtmp - mintmp)\n",
    "        scale_range = 1.0 + sigma/255.0/2.0\n",
    "        scale_shift = (1 - scale_range) / 2.0\n",
    "        xtilde = xtilde * scale_range + scale_shift\n",
    "        r = f(xtilde.float())\n",
    "        z= xtilde - r\n",
    "        z= (z - scale_shift) / scale_range\n",
    "        z = z * (maxtmp - mintmp) + mintmp\n",
    "        z = (1-beta)*f_est+beta*z\n",
    "        grad =  calculate_grad(f_est, degradation_mode, FB, FBC, FBFy, sf=Sf)\n",
    "        v_est = f_est - alpha*grad/sigma\n",
    "        w = (1-mu)*z+mu*v_est\n",
    "\n",
    "        residual[i] = torch.norm(w-x0, p=2)\n",
    "        x0 = w\n",
    "        ########\n",
    "        obj_fun[i] = calulate_data_term(k_tensor,degradation_mode, Sf,x0,y).float().cpu().numpy()\n",
    "    return x0, obj_fun, residual.float().cpu().numpy()\n",
    "\n",
    "def tensor_to_mat(imgname,img_name, x1, setting='DnCNN'):\n",
    "    x1_out = tensor2float(x1)\n",
    "    x_est_luma = single2uint(x1_out)\n",
    "    IR_filename = 'PSNR_mat/'+img_name+setting+str(noise_level)+'_Gaussian_blur_img_mat/'\n",
    "#     gt_filename = 'PSNR_mat/gt_mat/'\n",
    "    if not os.path.exists(IR_filename):\n",
    "        mkdir(IR_filename)\n",
    "    scio.savemat(IR_filename+imgname+'_luma.mat', {'data':x_est_luma})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "15526811",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "parrots - PSNR: 31.2551 dB\n"
     ]
    }
   ],
   "source": [
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "init_im1 = rgb2ycbcr(norm_proj(init_im))\n",
    "init_im2 = np.expand_dims(init_im1, axis=2)\n",
    "img_tensor = array2tensor(init_im2).to(device)\n",
    "if img_type == 'RGB' and Sf==1:\n",
    "    degradation_mode = 'deblurring'\n",
    "    x0 = img_tensor#初值\n",
    "elif Sf>1:\n",
    "    degradation_mode = 'SR'\n",
    "    x0 = cv2.resize(init_im1, (init_im1.shape[1] * Sf, init_im1.shape[0] * Sf),interpolation=cv2.INTER_CUBIC)\n",
    "    x0 = np.expand_dims(x0, axis=2)\n",
    "    x0 = utils_sr.shift_pixel(x0, Sf)\n",
    "#     print(x0.shape)\n",
    "    x0 = array2tensor(x0).to(device)\n",
    "else:\n",
    "    img_tensor = array2tensor(init_im).to(device)\n",
    "#     print(init_im2.dtype)\n",
    "#     print(np.max(init_im2))\n",
    "FB, FBC, F2B, FBFy, k_tensor = initialize_prox(img_tensor, degradation_mode, k, Sf, device)\n",
    "################################################algorithm setting\n",
    "K = 200\n",
    "beta = 0.26\n",
    "alpha = 32\n",
    "if noise_level<=5.0:\n",
    "    sigma_f=5\n",
    "elif 15 >= noise_level>5:\n",
    "    sigma_f=15\n",
    "elif 40 >= noise_level> 15:\n",
    "    sigma_f=25\n",
    "else:\n",
    "    print('error')\n",
    "model_type = 'DnCNN' #'RealSN_DnCNN'|'RealSN_SimpleCNN'\n",
    "# opt_RED={'alpha':2.4, 'beta':0.01, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'lambda':0.02,'input_sigma':noise_level**2, 'sf': Sf} #for sigma**2=2, starfish, uniform debluring\n",
    "opt_RED={'alpha':2.4, 'beta':0.01, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'lambda':0.02,'input_sigma':noise_level**2, 'sf': Sf} #for butterfly \n",
    "# opt_REDPRO={'alpha':2.4, 'beta':0.01, 'sigma':noise_level**2, 'sigma_f':sigma_f, 'K':2000, 'lambda':0.01,'input_sigma':noise_level**2, 'sf': Sf}# for parrots sigma=5\n",
    "opt_REDPRO={'alpha':2.4, 'beta':0.01, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'lambda':0.01,'input_sigma':noise_level**2, 'sf': Sf}\n",
    "opt_PnP_FBS={'alpha':2.4, 'beta':0.01, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'lambda':0.01,'input_sigma':noise_level**2, 'sf': Sf}\n",
    "# opt={'alpha':39, 'beta':0.27, 'sigma':5.0, 'sigma_f':5.0, 'K':500, 'mu_0':15} for RealSN_DnCNN, set12,sigma=2, uniform debluring\n",
    "opt={'alpha':4, 'beta':0.01, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':3000, 'mu_0':1500, 'sf': Sf}\n",
    "opt2={'alpha':2.4, 'beta':0.1, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':800, 'mu_0':4, 'sf': Sf} #for sigma**2=2, starfish, uniform debluring\n",
    "# opt={'alpha':43, 'beta':0.29, 'sigma':5.0, 'sigma_f':5.0, 'K':400, 'mu_0':15}\n",
    "opt_r={'alpha':2, 'beta':0.05, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'mu_0':500, 'sf': Sf}# for parrots(SR, sigma=5)\n",
    "# opt_r={'alpha':2, 'beta':0.1, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'mu_0':500, 'sf': Sf}\n",
    "# opt_r={'alpha':2.4, 'beta':0.01, 'sigma':noise_level, 'sigma_f':sigma_f, 'K':2000, 'mu_0':400, 'sf': Sf}\n",
    "f = load_model(model_type, int(opt['sigma_f']))\n",
    "with torch.no_grad():\n",
    "    if algorithm == 'RED':\n",
    "        x1, objfun,r = RED(x0,img_tensor, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt_RED)\n",
    "    elif algorithm == 'RED_PRO':\n",
    "        x1, objfun,r = REDPRO(x0,img_tensor, k_tensor, degradation_mode, FB, FBC, FBFy, f,  opt_REDPRO)\n",
    "    elif algorithm == 'SAM_PROv2':\n",
    "        x1, objfun,r = SAM_PRO_v2(x0,img_tensor, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt)\n",
    "    elif algorithm == 'PnP_FBS':\n",
    "        x1, objfun,r = PnP_FBS(x0,img_tensor, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt_PnP_FBS)\n",
    "    elif algorithm == 'SAM_PROv1':\n",
    "        x1, objfun,r = SAM_PRO_v1(x0,img_tensor, k_tensor, degradation_mode, FB, FBC, FBFy, f, opt_r)\n",
    "    else:\n",
    "        print('algorithm not implemented ^_^')\n",
    "eng = matlab.engine.start_matlab()\n",
    "eng.addpath(eng.genpath(eng.fullfile(os.getcwd(),'Metrics')))\n",
    "x_gt_luma = rgb2ycbcr(input_im_uint)\n",
    "im_ycbr = rgb2ycbcr(norm_proj(init_im), only_y=False)\n",
    "im_ycbr2 = rgb2ycbcr(norm_proj(init_im), only_y=False)\n",
    "x1_out = tensor2float(x1)\n",
    "x_est_luma = single2uint(x1_out)\n",
    "H, W, _ = input_im_uint.shape\n",
    "h, w = x_est_luma.shape\n",
    "x_est_luma = x_est_luma[1:h-1,1:w-1] if W==256 and H==256 else x_est_luma########keep the same size\n",
    "PSNR = eng.ComputePSNR(matlab.uint8(x_gt_luma.tolist()), matlab.uint8(x_est_luma.tolist()))###Fllowed by RED paper\n",
    "print('{} - PSNR: {:.4f} dB'.format(name, PSNR))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "49d0dac7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy.io as scio\n",
    "if Sf>1: ####256x256\n",
    "    scio.savemat('result_1.mat', {'data':im_ycbr})\n",
    "    scio.savemat('result_r.mat', {'data':im_ycbr2})\n",
    "    scio.savemat('degraded_img.mat', {'data':np.uint8(norm_proj(init_im)*255)})\n",
    "    init_im_cubic = MATLAB_imresize('degraded_img.mat', Sf)\n",
    "#         print(init_im_cubic.shape)\n",
    "    im_ycbr = MATLAB_imresize('result_1.mat', Sf)\n",
    "    im_ycbr[:,:, 0] = x1_out\n",
    "    im_ycbr = im_ycbr[1:h-1,1:w-1,0:3] if W==256 and H==256 else im_ycbr\n",
    "    init_im_cubic = init_im_cubic[1:h-1,1:w-1,0:3] if W==256 and H==256 else init_im_cubic\n",
    "    psnr0 = calculate_psnr(input_im_uint,init_im_cubic)\n",
    "else:\n",
    "    im_ycbr[:,:, 0] = x1_out\n",
    "    psnr0 = calculate_psnr(input_im_uint, np.uint8(norm_proj(init_im)*255))\n",
    "psnr_luma = calculate_psnr(x_gt_luma,x_est_luma)#####matlab psnr function\n",
    "\n",
    "x1_out = ycbcr2rgb(im_ycbr)\n",
    "x1_out = single2uint(x1_out)\n",
    "if Sf>1:\n",
    "    save_path = 'results/'+str(noise_level)+'_SR/'\n",
    "    if not os.path.exists(save_path):\n",
    "        mkdir(save_path)\n",
    "    save_IR_name = save_path+name+'_'+task+'_'+algorithm+'_IR.png'\n",
    "    save_degraded_name = save_path+name+'_deg.png'\n",
    "    save_degraded_name2 = save_path+name+'_deg2.png'\n",
    "    imsave(x1_out,save_IR_name)\n",
    "    imsave(np.uint8(norm_proj(init_im)*255),save_degraded_name)\n",
    "    img_cubic = MATLAB_imresize2(save_degraded_name, Sf)\n",
    "    imsave(img_cubic,save_degraded_name2)\n",
    "fig, ax = plt.subplots(nrows=1, ncols=3, figsize=(10*1, 16*3))\n",
    "ax[0].imshow(input_im)\n",
    "ax[0].set_title('Clean image')\n",
    "ax[0].set_axis_off()\n",
    "ax[1].imshow(np.uint8(norm_proj(init_im)*255))\n",
    "ax[1].set_title('Blur image')\n",
    "ax[1].set_axis_off()\n",
    "ax[2].imshow(x1_out)\n",
    "ax[2].set_title('Recovered')\n",
    "ax[2].set_axis_off()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "DL_img",
   "language": "python",
   "name": "dl_img"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}