LayerUtility: Check Mask

Documentation

• Class name: LayerUtility: CheckMask
• Category: 😺dzNodes/LayerUtility
• Output node: False

The CheckMask node is designed to evaluate the effectiveness of a given mask by determining if the area of the mask that meets certain criteria (e.g., white point value) exceeds a specified percentage. This functionality is crucial for filtering out ineffective masks in image processing workflows, ensuring that subsequent operations are performed on valid data.

Input types

Required

• mask
  • The mask input is the primary data on which the node operates, used to determine if the mask is effective based on its white point value and area coverage.
  • Comfy dtype: MASK
  • Python dtype: torch.Tensor
• white_point
  • The white point parameter sets the threshold value for considering a pixel as part of the effective area of the mask, playing a critical role in the mask's evaluation.
  • Comfy dtype: INT
  • Python dtype: int
• area_percent
  • The area percent parameter specifies the minimum percentage of the mask that must be considered effective, based on the white point criteria, for the mask to be deemed valid.
  • Comfy dtype: INT
  • Python dtype: int

Optional

Output types

• bool
  • Comfy dtype: BOOLEAN
  • Indicates whether the mask is considered effective based on the white point and area percentage criteria.
  • Python dtype: bool

Usage tips

• Infra type: GPU
• Common nodes: unknown

Source code

class CheckMask:

    def __init__(self):
        pass

    @classmethod
    def INPUT_TYPES(self):
        blank_mask_list = ['white', 'black']
        return {
            "required": {
                "mask": ("MASK",),  #
                "white_point": ("INT", {"default": 1, "min": 1, "max": 254, "step": 1}), # 用于判断mask是否有效的白点值，高于此值被计入有效
                "area_percent": ("INT", {"default": 1, "min": 1, "max": 99, "step": 1}), # 区域百分比，低于此则mask判定无效
            },
            "optional": { #
            }
        }

    RETURN_TYPES = ("BOOLEAN",)
    RETURN_NAMES = ('bool',)
    FUNCTION = 'check_mask'
    CATEGORY = '😺dzNodes/LayerUtility'

    def check_mask(self, mask, white_point, area_percent,):

        if mask.dim() == 2:
            mask = torch.unsqueeze(mask, 0)
        mask = tensor2pil(mask[0])
        if mask.width * mask.height > 262144:
            target_width = 512
            target_height = int(target_width * mask.height / mask.width)
            mask = mask.resize((target_width, target_height), Image.LANCZOS)
        return (mask_white_area(mask, white_point) * 100 > area_percent,)