Class _ClassUniform

transforms/joint_transforms.py:579–712

class _ClassUniform(object):
    def __init__(self, size, crop_nopad, scale_min=0.5, scale_max=2.0, ignore_index=0,
                 class_list=[16, 15, 14]):
        """
        This is the initialization for class uniform sampling
        :param size: crop size (int)
        :param crop_nopad: Padding or no padding (bool)
        :param scale_min: Minimum Scale (float)
        :param scale_max: Maximum Scale (float)
        :param ignore_index: The index value to ignore in the GT images (unsigned int)
        :param class_list: A list of class to sample around, by default Truck, train, bus
        """
        self.size = size
        self.crop = RandomCrop(self.size, ignore_index=ignore_index, nopad=crop_nopad)

        self.class_list = class_list.replace(" ", "").split(",")

        self.scale_min = scale_min
        self.scale_max = scale_max

    def detect_peaks(self, image):
        """
        Takes an image and detect the peaks usingthe local maximum filter.
        Returns a boolean mask of the peaks (i.e. 1 when
        the pixel's value is the neighborhood maximum, 0 otherwise)

        :param image: An 2d input images
        :return: Binary output images of the same size as input with pixel value equal
        to 1 indicating that there is peak at that point
        """

        # define an 8-connected neighborhood
        neighborhood = generate_binary_structure(2, 2)

        # apply the local maximum filter; all pixel of maximal value
        # in their neighborhood are set to 1
        local_max = maximum_filter(image, footprint=neighborhood) == image
        # local_max is a mask that contains the peaks we are
        # looking for, but also the background.
        # In order to isolate the peaks we must remove the background from the mask.

        # we create the mask of the background
        background = (image == 0)

        # a little technicality: we must erode the background in order to
        # successfully subtract it form local_max, otherwise a line will
        # appear along the background border (artifact of the local maximum filter)
        eroded_background = binary_erosion(background, structure=neighborhood,
                                           border_value=1)

        # we obtain the final mask, containing only peaks,
        # by removing the background from the local_max mask (xor operation)
        detected_peaks = local_max ^ eroded_background

        return detected_peaks

    def __call__(self, img, mask):
        """