update LAB6

import numpy as np

import numpy.fft

from skimage import io, data

import math

from scipy import interpolate

import matplotlib.pyplot as plt

from matplotlib import cm

import matplotlib.image as mplimg

import EE326_SUSTech as ee

def adaptive_11810818(input_image, n_size, smax):

    input_image = io.imread(input_image + ".tiff")

    output_image = np.zeros(input_image.shape, dtype=np.uint8)

    m, n = input_image.shape

    for i in range(m):

        for j in range(n):

            n_size_2 = n_size

            while True:

                step = (int)((n_size_2 - 1) / 2)

                pixels = np.zeros(n_size_2 * n_size_2)

                for i2 in range(n_size_2):

                    for j2 in range(n_size_2):

                        if i - step + i2 >= 0 \

                                and i - step + i2 < input_image.shape[0] \

                                and j - step + j2 >= 0 \

                                and j - step + j2 < input_image.shape[0]:

                            pixels[j2 * n_size_2 + i2] = input_image[i - step + i2, j - step + j2]

                pixels_sorted = np.sort(pixels)

                med = (int)((n_size_2 * n_size_2-1)/2)

                a1 = pixels_sorted[med] - pixels_sorted[0]

                a2 = pixels_sorted[med] - pixels_sorted[n_size_2 * n_size_2-1]

                if(a1>0 and a2<0):

                    b1 = input_image[i, j] - pixels_sorted[0]

                    b2 = input_image[i, j] - pixels_sorted[n_size_2 * n_size_2-1]

                    if(b1>0 and b2<0):

                        output_image[i, j] = pixels[med]

                    else:

                        output_image[i, j] = pixels_sorted[med]

                    break

                else:

                    if(n_size_2 < smax):

                        n_size_2 += 2

                    else:

                        output_image[i, j] = pixels_sorted[med]

                        break

    return output_image

if __name__ == '__main__':

    for i in [1

        , 2

        , 3

        , 4

              ]:

        input_image = "Q6_1_" + str(i)

        output_name = "plots/" + str(input_image) + "_adaptive.png"

        output_image = adaptive_11810818(input_image, 3, 20)

        mplimg.imsave(output_name,

                      output_image,

                      cmap=cm.gray)

        print("Finish processing " + str(i))

 No newline at end of file

import numpy as np

import numpy.fft

from skimage import io, data

import math

from scipy import interpolate

import matplotlib.pyplot as plt

from matplotlib import cm

import matplotlib.image as mplimg

import EE326_SUSTech as ee

def denoise_11810818(input_image, n_size, mode):

    output_name = "plots/" + str(input_image) + "_denoised.png"

    input_image = io.imread(input_image+".tiff")

    for i in mode:

        output_image = ee.denoise_filter(input_image, n_size, i)

    mplimg.imsave(output_name,

                  output_image,

                  cmap=cm.gray)

if __name__ == '__main__':

    # denoise_11810818("Q6_1_1", 3, ["medium"])

    # denoise_11810818("Q6_1_2", 5, ["medium"])

    denoise_11810818("Q6_1_3", 9, ["medium"])

    # denoise_11810818("Q6_1_4", 9, ["medium"])