# detect sharp edges and produce binary output
# think of it as high pass filtering operation ->
# it allows high frequency content to pass through and blocks
# the low frequency content
# build a kernel of a high pass filter
# we will use sobel filter, whose kernel is
# Sx = np.array([[-1,0,1],[-2,0,2],[-1,0,1]])
# Sy = np.array([[-1,-2,-1],[0,0,0],[1,2,1]])
import cv2
import numpy as np
import myLibrary
image = cv2.imread('image.jpg',cv2.IMREAD_GRAYSCALE)
newimage = myLibrary.resize(image,500)
# get the shape (height and width)
rows,cols = newimage.shape
sobel_h = cv2.Sobel(newimage, cv2.CV_64F, 1, 0, ksize=5)
sobel_v = cv2.Sobel(newimage, cv2.CV_64F, 0, 1, ksize=5)
cv2.imshow('Original Image', newimage)
cv2.imshow('Sobel Horizontal', sobel_h)
cv2.imshow('Sobel Vertical', sobel_v)
# you can also use laplacian filter
laplacian = cv2.Laplacian(newimage,cv2.CV_64F)
cv2.imshow('Laplacian filter',laplacian)
# laplacian give rise to noise in the picture
# to overcome that we can use Canny edge detector
# a =[]
image = cv2.imread('testimage.jpg')
myLibrary.showOriginal(image)
cv2.waitKey(0)
print()
# # time.sleep(2)
print("Demonstration Translation (shifting of an object along x and y axis) ")
n = input("Enter x and y (seperated by space): ")
val = n.split(' ')
a = int(val[0])
b = int(val[1])
translatedImage = myLibrary.translate(image,a,b)
cv2.imshow("Translated Image",translatedImage)
cv2.waitKey(0)
print()
print("Demonstration of Rotation by an angle")
n = int(input("Enter an angle "))
rotatedImage = myLibrary.rotate(image,n)
cv2.imshow("Rotated Image",rotatedImage)
cv2.waitKey(0)
print()
print("Demonstration of Resizing")
n = input("Enter new width and height (seperated by space): ")
val = n.split(' ')
width = int(val[0])
height = int(val[1])
resizedImage = myLibrary.resize(image,width,height)
cv2.imshow("resizedImage",resizedImage)
cv2.waitKey(0)