def prewitt(img):
filter_x = np.array([[-1, 0, 1], [-1, 0, 1], [-1, 0, 1]])
filter_y = np.array([[-1, -1, -1], [0, 0, 0], [1, 1, 1]])
img_x = conv_img(img, filter_x)
img_y = conv_img(img, filter_y)
sq_img = np.sqrt(img_x**2 + img_y**2)
img_out = np.floor(sq_img * (256 / np.max(sq_img)))
return img_out
def filter_gauss(img):
gauss_filter = np.asarray([[0, 1, 2, 1, 0], [1, 3, 5, 3,
1], [2, 5, 9, 5, 2],
[1, 3, 5, 3, 1], [0, 1, 2, 1, 0]])
gauss_filter = gauss_filter / np.sum(gauss_filter, axis=None)
img_conv = conv_img(img, gauss_filter)
return img_conv
def LOG(img):
filter_LOG = np.array([[-2, -4, -4, -4, -2], [-4, 0, 8, 0, -4],
[-8, 8, 24, 8, -4], [-4, 0, 8, 0, -4],
[-2, -4, -4, -4, -2]])
img_conv = conv_img(img, filter_LOG)
abs_img = np.abs(img_conv / np.sum(filter_LOG[filter_LOG > 0]))
img_out = np.floor(abs_img).astype(int)
return img_out
def laplace(img, alpha=0.4):
filter_laplace = np.array([[-1, -1, -1], [-1, 8, -1], [-1, -1, -1]])
img_conv = conv_img(img, filter_laplace)
# abs_img = np.abs(img_conv)
img_nopad = img_conv / np.max(filter_laplace)
img_pad = np.pad(img_nopad, [(1, 1), (1, 1)],
mode='constant') # 2*pad = 2*1 = (h of filter - 1) = 3-1
return regularize(alpha * img_pad + img)
def filter_box(img):
box_filter = np.ones((7, 7)) / (7 * 7)
img_conv = conv_img(img, box_filter)
return img_conv
def filter_laplace(img):
laplace_filter = (1.0 / 16) * np.array([[0, 0, -1, 0, 0], [
0, -1, -2, -1, 0
], [-1, -2, 16, -2, -1], [0, -1, -2, -1, 0], [0, 0, -1, 0, 0]])
img_conv = conv_img(img, laplace_filter)
return img_conv