def robertsFilters(image):
"""
Apply Roberts filters in X and Y directions
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
colourers.info(f'Applying Roberts filter in X and Y directions')
Kx = np.array([[1, 0], [0, -1]], np.float32)
Ky = np.array([[0, 1], [-1, 0]], np.float32)
Ix = optimizedConv2D(image, Kx)
Iy = optimizedConv2D(image, Ky)
G = np.hypot(Ix, Iy)
G = G / G.max() * 255
theta = np.arctan2(Iy, Ix)
return (G, theta)
def kirschFilters(image):
"""
Apply Kirsch filters in X and Y directions
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
colourers.info(f'Applying Kirsch filter in X and Y directions')
Kx = np.array([[-3, -3, 5], [-3, 0, 5], [-3, -3, 5]], np.float32)
Ky = np.array([[-3, -3, -3], [-3, 0, -3], [5, 5, 5]], np.float32)
Ix = optimizedConv2D(image, Kx)
Iy = optimizedConv2D(image, Ky)
G = np.hypot(Ix, Iy)
G = G / G.max() * 255
theta = np.arctan2(Iy, Ix)
return (G, theta)
def gaborFilter(image, kernel):
"""
Performing a gabor filter
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
return optimizedConv2D(image, kernel).astype(np.uint8)
def averageBlur(image):
"""
Apply an average blur on the image
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
averageBlurKernel = np.ones((5, 5)) * 1.0 / 25
return optimizedConv2D(image, averageBlurKernel)
def simpleBlur(image):
"""
Apply a simple blur on the image
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
simpleBlurKernel = np.ones((3, 3)) * 1.0 / 9
return optimizedConv2D(image, kernel=simpleBlurKernel)
def increasedEdgeEnhancement(image):
"""
Performing an increased edge enhancement filter
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
increasedEdgeEnhanceKernel = np.array(
[[-1, -1, -1], [-1, 9, -1], [-1, -1, -1]], np.float32)
return optimizedConv2D(image, increasedEdgeEnhanceKernel).clip(0, 255)
def sharpen(image):
"""
Performing a sharpening filter
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
increasedEdgeEnhanceKernel = np.array(
[[0, -1, 0], [-1, 5, -1], [0, -1, 0]], np.float32)
return optimizedConv2D(image, increasedEdgeEnhanceKernel).clip(0, 255)
def unsharp(image):
"""
Performing an unsharpening filter
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
increasedEdgeEnhanceKernel = np.array(
[[1, 4, 6, 4, 1], [4, 16, 24, 16, 4], [6, 24, -476, 24, 6],
[4, 16, 24, 16, 4], [1, 4, 6, 4, 1]], np.float32) * -1.0 / 256
return optimizedConv2D(image, increasedEdgeEnhanceKernel).clip(0, 255)
def emboss(image):
"""
Performing an emboss filter
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
embossKernel = np.array([
[-2, -1, 0],
[-1, 1, 1],
[ 0, 1, 2]
])
return optimizedConv2D(image, embossKernel).clip(0, 255)
def kirschEdgeDetection(image, sigma=1, kernelSize=5):
"""
Apply the Kirsch filter for an edge detection
Parameters
----------
image: ndarray((h, w, 3))
sigma: float
kernelSize: int
Returns
-------
ndarray((h, w, 3))
"""
smoothedImage = optimizedConv2D(image, gaussianKernel(kernelSize, sigma))
gradientMatrix, thetaMatrix = kirschFilters(smoothedImage)
return gradientMatrix
def motionBlur(image):
"""
Apply a motion blur on the image
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
motionBlurKernel = np.array([
[1, 0, 0, 0, 1],
[0, 1, 0, 1, 0],
[0, 0, 1, 0, 0],
[0, 1, 0, 1, 0],
[1, 0, 0, 0, 1]
]) * 1.0 / 9
return optimizedConv2D(image, motionBlurKernel)
def gaussianBlur(image):
"""
Apply a gaussian blur on the image
Parameters
----------
image: ndarray((h, w, 3))
Returns
-------
ndarray((h, w, 3))
"""
gaussianBlurKernel = np.array([
[2, 4, 5, 4, 2],
[4, 9, 12, 9, 4],
[5, 12, 15, 12, 5],
[4, 9, 12, 9, 4],
[2, 4, 5, 4, 2]
]) * 1.0 / 159
return optimizedConv2D(image, gaussianBlurKernel)
def cannyEdgeDetection(image,
sigma=1,
kernelSize=5,
weakPix=75,
strongPix=255,
lowThreshold=0.05,
highThreshold=0.15):
"""
Apply the Canny filter for an edge detection
Parameters
----------
image: ndarray((h, w, 3))
sigma: float
kernelSize: int
weakPix: int
strongPix: int
lowThreshold: int
highThreshold: int
Returns
-------
ndarray((h, w, 3))
"""
smoothedImage = optimizedConv2D(image, gaussianKernel(kernelSize, sigma))
gradientMatrix, thetaMatrix = sobelFilters(smoothedImage)
nonMaxImage = nonMaxSuppression(gradientMatrix, thetaMatrix)
thresholdImage = threshold(nonMaxImage,
lowThresholdRatio=lowThreshold,
highThresholdRatio=highThreshold,
weakPix=weakPix,
strongPix=strongPix)
return hysteresis(thresholdImage, weakPixel=weakPix, strongPixel=strongPix)
