forensicfit.utils.image_tools module#
- forensicfit.utils.image_tools.rotate_image(image, angle)[source]#
Rotates the image by angle degrees
- Parameters:
angle (float) – Angle of rotation.
- Returns:
None.
- forensicfit.utils.image_tools.gaussian_blur(image, window=(15, 15))[source]#
This method applies Gaussian Blur filter to the image.
- Parameters:
window (tuple int, optional) – The window in which the gaussian blur is going to be applied. The default is (15,15).
- Return type:
None.
- forensicfit.utils.image_tools.split_v(image, pixel_index=None, pick_side='L', flip=True)[source]#
This method splits the image in 2 images based on the fraction that is given in pixel_index
- Parameters:
pixel_index (float, optional) – fraction in which the image is going to be split. The value should be a number between zero and one. The default is 0.5.
pick_side (str, optional) – The side in which will over write the image in the class. The default is ‘L’.
- Return type:
None.
- forensicfit.utils.image_tools.to_gray(image, mode='SD')[source]#
Gray Scale image of the input image.
modes: ‘BT.470’ and ‘BT.709’ SD ‘BT.470’ : Y = 0.299 R + 0.587 G + 0.114 B HD ‘BT.709’ : Y = 0.2125 R + 0.7154 G + 0.0721 B
- Return type:
ndarray- Returns:
gray_scale – Gray Scale image of the input image.
- Return type:
cv2 object
- forensicfit.utils.image_tools.contours(image, mask_threshold=60)[source]#
A list of pixels that create the contours in the image
- Returns:
contours – A list of pixels that create the contours in the image
- Return type:
list
- forensicfit.utils.image_tools.largest_contour(contours)[source]#
A list of pixels forming the contour with the largest area
- Returns:
contour_max_area – A list of pixels forming the contour with the largest area
- Return type:
list
- forensicfit.utils.image_tools.remove_background(image, contour, outside=True, pixel_value=0)[source]#
Removes the background outside or inside the contour
- Parameters:
image (np.array) – _description_
contour (np.array) – _description_
outside (bool, optional) – _description_, by default True
pixel_value (int, optional) – _description_, by default 0
- forensicfit.utils.image_tools.get_masked(image, mask_threshold)[source]#
Populates the masked image with the gray scale threshold Returns
- forensicfit.utils.image_tools.resize(image, size)[source]#
This method resizes the image to the pixel size given.
- Parameters:
size (tuple int,) – The target size in which the image is going to be resized.
- Return type:
None.
- forensicfit.utils.image_tools.exposure_control(image, mode='equalize_hist', **kwargs)[source]#
modifies the exposure
- Return type:
ndarray- Parameters:
mode (str, optional) – Type of exposure correction. It can be selected from the options:
'equalize_hist'or'equalize_adapthist'. equalize_hist and equalize_adapthist <https://scikit-image.org/docs/stable/api/skimage.exposure.html#equalize-adapthist> use sk-image. by default ‘equalize_hist’
- forensicfit.utils.image_tools.apply_filter(image, mode, **kwargs)[source]#
Applies different types of filters to the image
- Return type:
ndarray- Parameters:
mode (str) – Type of filter to be applied. The options are *
'meijering': <Meijering neuriteness filter https://scikit-image.org/docs/stable/api/skimage.filters.html#skimage.filters.meijering>_, *'frangi': < Frangi vesselness filter https://scikit-image.org/docs/stable/api/skimage.filters.html#skimage.filters.frangi>_, *'prewitt': <Prewitt transform https://scikit-image.org/docs/stable/api/skimage.filters.html#prewitt>_, *'sobel': <Sobel filter https://scikit-image.org/docs/stable/api/skimage.filters.html#skimage.filters.sobel>_, *'scharr': <Scharr transform https://scikit-image.org/docs/stable/api/skimage.filters.html#skimage.filters.scharr>, *'roberts': <Roberts’ Cross operator https://scikit-image.org/docs/stable/api/skimage.filters.html#examples-using-skimage-filters-roberts>_, *'sato': <Sato tubeness filter https://scikit-image.org/docs/stable/api/skimage.filters.html#skimage.filters.sato>_.