-
Notifications
You must be signed in to change notification settings - Fork 1
PyPlaque_utils_segment_plaque
PyPlaque.utils.segment_plaque
get_all_plaque_regions Function
This function identifies and labels all connected regions in a binary image that are likely to
contain virus plaques. It processes the input grayscale image by applying a threshold to
create a binary mask, then computes the distance transform of the inverted binary mask to
identify plaque candidates. Connected components within this distance-transformed image are
labeled and returned as a label matrix. The original background pixels in the label matrix are
set to 0 where they were not present in the input binary mask.
Args: image (np.ndarray, required): A 2D numpy array representing the grayscale image of the tissue section. threshold (float, required): A float value that determines the pixel intensity below which pixels are considered part of the background. plq_connect (int, required): An integer specifying the maximum distance within which connected components are grouped to be considered as potential virus plaques.
Returns: np.ndarray: A 2D numpy array of integers where each unique value represents a different connected region in the input image, likely containing virus plaques. Pixels not part of any plaque or background are set to 0.
Raises:
TypeError: If image is not a 2D numpy array, threshold is not a float, or plq_connect
is not an integer.
ValueError: If threshold is outside the valid range for pixel intensities in image or
if plq_connect is less than or equal to zero.
get_plaque_mask Function
This function generates a mask of virus plaques in an input image based on specified parameters.
It processes the input image to identify and segment regions that are likely to contain virus
plaques. It uses thresholding and morphological operations to create a binary mask where plaque
regions are white (1) and background is black (0). The function optionally performs fine-grained
plaque detection if enabled in virus_params.
Args: input_image (np.ndarray, required): A 2D numpy array representing the grayscale image of the tissue section. virus_params (dict, required): A dictionary containing parameters for virus plaque detection, including threshold value, connectivity, and other morphological operations settings.
Returns:
tuple: A tuple containing two elements:
- final_plq_reg_image (np.ndarray): A 2D numpy array of the same size as input_image
with plaque regions marked by white pixels (1) and background by black pixels (0).
- global_peak_coords (np.ndarray or None): An array of coordinates where local peaks
were detected in the final mask, if fine detection is enabled; otherwise, returns None.
Raises:
TypeError: If input_image is not a 2D numpy array or virus_params is not a dictionary.
ValueError: If any parameter within virus_params does not match its expected type or value
range as specified in the method signature.
plot_virus_contours Function This function plots contours of virus plaques on a modified grayscale image. It processes an input image to remove its background and then generates a mask for the plaque region. It uses these masks to find and plot the contours of the virus plaques using custom colors and markers. The final image is displayed interactively or saved to disk if a save path is provided.
Args:
input_image (np.ndarray, required): A 2D numpy array representing the grayscale input image
of the tissue section containing virus plaques.
virus_params (dict, required): A dictionary containing parameters for virus plaque detection
and correction, including 'correction_ball_radius'.
save_path (str or None, optional): The file path where the plot will be saved if provided;
otherwise, it is displayed interactively. Defaults to None.
Returns: None: The function generates a matplotlib plot based on the arguments provided and optionally saves it to disk.
Raises: TypeError: If any of the input arguments do not match their expected types as specified in the method signature.