main.py

import cv2
import numpy as np
from matplotlib import pyplot as plt

from methods import Bilinear, HQL, HA
from methods import GBTF, DLMMSE
from methods import RI, IRI, MLRI

def metrics(raw_img, new_img):
    result = []
    for c in range(0, 3):
        mse = np.mean(np.power((raw_img[5:-5, 5:-5, c] - new_img[5:-5, 5:-5, c]), 2))
        if mse == 0:
            result.append('perfect!!')
        else:
            result.append(10 * np.log10(255*255/mse))

    return result

def make_bayer(img):
    new_img = np.zeros_like(img)

    new_img[0::2, 0::2, 0] = img[0::2, 0::2, 0]
    new_img[0::2, 1::2, 1] = img[0::2, 1::2, 1]
    new_img[1::2, 0::2, 1] = img[1::2, 0::2, 1]
    new_img[1::2, 1::2, 2] = img[1::2, 1::2, 2]

    return new_img

for picname in ['./kodim19.png']:
    src_img = cv2.imread(picname)
    src_img = cv2.cvtColor(src_img, cv2.COLOR_BGR2RGB)

    bayer_img = make_bayer(src_img)
    plt.imshow(bayer_img), plt.show()

    bilinear_img = Bilinear.run(bayer_img)
    plt.imshow(bilinear_img), plt.show()
    print('Bilinear: ', metrics(src_img, bilinear_img))

    hql_img = HQL.run(bayer_img)
    plt.imshow(hql_img), plt.show()
    print('HQL: ', metrics(src_img, hql_img))

    ha_img = HA.run(bayer_img)
    plt.imshow(ha_img), plt.show()
    print('HA: ', metrics(src_img, ha_img))

    dlmmse_img = DLMMSE.run(bayer_img)
    plt.imshow(dlmmse_img), plt.show()
    print('DLMMSE: ', metrics(src_img, dlmmse_img))

    gbtf_img = GBTF.run(bayer_img)
    plt.imshow(gbtf_img), plt.show()
    print('GBTF: ', metrics(src_img, gbtf_img))

    ri_img = RI.run(bayer_img)
    plt.imshow(ri_img), plt.show()
    print('RI: ', metrics(src_img, ri_img))

    mlri_img = MLRI.run(bayer_img)
    plt.imshow(mlri_img), plt.show()
    print('MLRI: ', metrics(src_img, mlri_img))

    iri_img = IRI.run(bayer_img)
    plt.imshow(iri_img), plt.show()
    print('IRI: ', metrics(src_img, iri_img))