vardanagarwal

from tensorflow.keras.models import Model
from tensorflow.keras.layers import SeparableConv2D
from tensorflow.keras.layers import MaxPooling2D
from tensorflow.keras.layers import BatchNormalization
from tensorflow.keras.layers import SpatialDropout2D, GaussianDropout, Dropout, ActivityRegularization
from tensorflow.keras.layers import Flatten, Dense
from tensorflow.keras.layers import Input
from tensorflow.keras.layers import Add, concatenate
from tensorflow.keras.utils import plot_model

vardanagarwal

import cv2

img = cv2.imread('Paris.jpg')
dst_gray, dst_color = cv2.pencilSketch(img, sigma_s=60, sigma_r=0.07, shade_factor=0.05) # inbuilt function to generate pencil sketch in both color and grayscale
# sigma_s controls the size of the neighborhood. Range 1 - 200
# sigma_r controls the how dissimilar colors within the neighborhood will be averaged. A larger sigma_r results in large regions of constant color. Range 0 - 1
# shade_factor is a simple scaling of the output image intensity. The higher the value, the brighter is the result. Range 0 - 0.1
cv2.imshow("Image", img)
cv2.imshow("Output2", dst_gray)
cv2.imshow("Output", dst_color)

vardanagarwal

import cv2
import numpy as np

img = cv2.imread('Paris.jpg')
height, width = img.shape[:2]
y = np.ones((height, width), np.uint8) * 128
output = np.zeros((height, width), np.uint8)
# generating the kernels
kernel1 = np.array([[0, -1, -1], # kernel for embossing bottom left side
                    [1, 0, -1],

vardanagarwal

import cv2

img = cv2.imread('Paris.jpg')
edges1 = cv2.bitwise_not(cv2.Canny(img, 100, 200)) # for thin edges and inverting the mask obatined
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
gray = cv2.medianBlur(gray, 5) # applying median blur with kernel size of 5
edges2 = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 7, 7) # thick edges
dst = cv2.edgePreservingFilter(img, flags=2, sigma_s=64, sigma_r=0.25) # you can also use bilateral filter but that is slow
# flag = 1 for RECURS_FILTER (Recursive Filtering) and 2 for  NORMCONV_FILTER (Normalized Convolution). NORMCONV_FILTER produces sharpening of the edges but is slower.
# sigma_s controls the size of the neighborhood. Range 1 - 200

vardanagarwal

import cv2
import numpy as np

img = cv2.imread('paris.jpg')
original = img.copy()
img = np.array(img, dtype=np.float64) # converting to float to prevent loss
img = cv2.transform(img, np.matrix([[0.272, 0.534, 0.131],
                                    [0.349, 0.686, 0.168],
                                    [0.393, 0.769, 0.189]])) # multipying image with special sepia matrix
img[np.where(img > 255)] = 255 # normalizing values greater than 255 to 255

vardanagarwal

import cv2

def exponential_function(channel, exp):
    table = np.array([min((i**exp), 255) for i in np.arange(0, 256)]).astype("uint8") # creating table for exponent
    channel = cv2.LUT(channel, table)
    return channel
def tone(img, number):
    for i in range(3):
        if i == number:
            img[:, :, i] = exponential_function(img[:, :, i], 1.05) # applying exponential function on slice

vardanagarwal

import cv2
import numpy as np

def hsv(img, l, u):
    hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
    lower = np.array([l,128,128]) # setting lower HSV value
    upper = np.array([u,255,255]) # setting upper HSV value
    mask = cv2.inRange(hsv, lower, upper) # generating mask
    return mask

vardanagarwal

import cv2
import numpy as np

img = cv2.imread('Paris.jpg')
original = img.copy()
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
xp = [0, 64, 112, 128, 144, 192, 255] # setting reference values
fp = [0, 16, 64, 128, 192, 240, 255] # setting values to be taken for reference values
x = np.arange(256)
table = np.interp(x, xp, fp).astype('uint8') # creating lookup table

vardanagarwal

import cv2
import numpy as np

img = cv2.imread('Paris.jpg')
height, width = img.shape[:2]
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
thresh = 0.8 # creating threshold. This means noise will be added to 80% pixels
for i in range(height):
    for j in range(width):
        if np.random.rand() <= thresh:

vardanagarwal

import cv2
import numpy as np

img = cv2.imread('Paris.jpg')
image_HLS = cv2.cvtColor(img,cv2.COLOR_BGR2HLS) # Conversion to HLS
image_HLS = np.array(image_HLS, dtype = np.float64)
daylight = 1.15
image_HLS[:,:,1] = image_HLS[:,:,1]*daylight # scale pixel values up for channel 1(Lightness)
image_HLS[:,:,1][image_HLS[:,:,1]>255]  = 255 # Sets all values above 255 to 255
image_HLS = np.array(image_HLS, dtype = np.uint8)