scripts/tap_const.py

from pathlib import Path

import cv2

parent = Path(__file__).parent.resolve()
root = parent.joinpath("..").joinpath("..").resolve()

datain = root.joinpath("datain").resolve()
images = datain.joinpath("images").resolve()
phenopsis = images.joinpath("phenopsis").resolve()
cvpp = images.joinpath("Plant_Phenotyping_Datasets").resolve()
param_set = images.joinpath("param_setting").resolve()
param_set_images = images.joinpath("param_setting").joinpath("images").resolve()
param_set_masks = images.joinpath("param_setting").joinpath("masks").resolve()
param_set_thresholds = images.joinpath("param_setting").joinpath("thresholds").resolve()
param_set_demo = images.joinpath("param_setting").joinpath("demo").resolve()
param_set_gt_seg = (
    images.joinpath("param_setting").joinpath("ground_truth_segmented").resolve()
)
param_set_train_data = images.joinpath("param_setting").joinpath("train_data").resolve()
exp_phenopsis = datain.joinpath("phenopsis").resolve()
data = datain.joinpath("data").resolve()
datasets = datain.joinpath("datasets").resolve()

dataout = root.joinpath("dataout").resolve()
configurations = dataout.joinpath("confs").resolve()
mask_cache = dataout.joinpath("masks").resolve()
masks_raw = dataout.joinpath("raw_masks").resolve()

checkpoints = root.joinpath("checkpoints").resolve()

notebooks = root.joinpath("notebooks")
scripts = notebooks.joinpath("scripts")

# MARK: Constants
DI_SRC = "source"
DI_SRC_BBOXES = "Source with bboxes"

DI_NONE = "no image"
DI_SEG_SEED_MASK = "segmentation seed mask"
DI_SEG_SEED_MASK_KEEP = "segmentation seed mask keep"
DI_SEG_SEED_MASK_DISCARD = "segmentation seed mask discard"
DI_SEG_DISTANCE_MAP = "segmentation distance map"
DI_SEG_DISTANCE_MAP_KEEP = "segmentation distance map keep"
DI_SEG_DISTANCE_MAP_DISCARD = "segmentation distance map discard"
DI_SEG_DISTANCE_MAP_KEEP_WITH_SEEDS = "segmentation distance map keep with seeds"
DI_SEG_DISTANCE_MAP_DISCARD_WITH_SEEDS = "segmentation distance map discard with seeds"
DI_SEG_SEEDS = "segmentation seeds"
DI_SEG_RAW_MASK = "segmentation raw mask"
DI_SEG_LOW_RES_LOGITS_MASK = "segmentation low resolution logits mask"
DI_SEG_EVOLVING_MASK = "segmentation mask progression"
DI_SEG_CLEAN_MASK = "segmentation clean mask"
DI_SEG_CLEAN_MASK_DEMO = "segmentation clean mask demo"
DI_SEG_CLEAN_MASK_ON_IMAGE = "segmentation clean mask on image"
DI_SEG_CONTOURS_ON_SRC_IMAGE = "contours on source image"
DI_SEG_CLEAN_MASK_ON_IMAGE_WITH_CONTOURS = "clean mask on image with contours"
DI_SEG_MASK_ON_IMAGE_SUSPECT_CONTOURS = "masked image with suspect contours"
DI_SEG_CLEAN_MASK_ON_IMAGE_WTH_BOXES = "masked image with boxes"
DI_SEG_CLEAN_MASK_ON_IMAGE_WITH_SEEDS = "segmentation clean mask on image with seeds"
DI_SEG_PLANT_ONLY = "segmentation plant only"

DI_SEQ_SRC_IMAGE = "source image"
DI_SEG_RM_MASK = "raw mask"
DI_SEQ_RM_PLANT = "plant only"
DI_SEQ_RM_PLANT_HIGHLIGHT = "plant highlight"
DF_SEQ_RM_CONTOURS = "plant with contours"
DI_SEQ_RM_CONTOURS_HIERARCHY = "plant with contour hierarchy"
DI_SEQ_RM_SUSPECT_CONTOURS = "plant and suspect contours"
DI_SEQ_SM_MASK = "sifted mask"
DI_SEQ_SM_PLANT = "sifted plant"
DI_SEQ_SM_CONTOURS = "sifted plant with contours"
DI_SEQ_SM_CONTOURS_ON_SRC = "source image with sifted contours"

DI_SYMP_SEED_MASK = "symptoms seed mask"
DI_SYMP_SEED_MASK_KEEP = "symptoms seed mask keep"
DI_SYMP_SEED_MASK_DISCARD = "symptoms seed mask discard"
DI_SYMP_SEEDS = "symptoms seeds"


C_BLACK = (0, 0, 0)
C_BLUE = (255, 0, 0)
C_BLUE_VIOLET = (226, 43, 138)
C_CABIN_BLUE = (209, 133, 67)
C_CYAN = (255, 255, 0)
C_DIM_GRAY = (105, 105, 105)
C_FUCHSIA = (255, 0, 255)
C_GREEN = (0, 128, 0)
C_LIGHT_STEEL_BLUE = (222, 196, 176)
C_LIME = (0, 255, 0)
C_MAROON = (0, 0, 128)
C_ORANGE = (80, 127, 255)
C_PURPLE = (128, 0, 128)
C_RED = (0, 0, 255)
C_SILVER = (192, 192, 192)
C_TEAL = (128, 128, 0)
C_WHITE = (255, 255, 255)
C_YELLOW = (0, 255, 255)
C_BROWN = (42, 42, 165)
C_SIENNA = (45, 82, 160)


KLC_FULLY_INSIDE = dict(val=0, color=C_GREEN)
KLC_OVERLAPS = dict(val=1, color=C_LIME)
KLC_OK_TOLERANCE = dict(val=4, color=C_TEAL)
KLC_SMALL = dict(val=5, color=C_BLUE)
KLC_FAR = dict(val=6, color=C_RED)
KLC_SMALL_FAR = dict(val=7, color=C_FUCHSIA)


MP_HULL = "masp_property_hull"
MP_CENTROID = "masp_property_centroid"
MP_BOUND_RECT = "masp_property_bounding_rectangle"
MP_ROT_BOUND_RECT = "masp_property_rotated_bounding_rectangle"
MP_ENCLOSING_CIRCLE = "masp_property_enclosing_circle"
AVAILABLE_MASK_PROPERTIES = [
    MP_HULL,
    MP_CENTROID,
    MP_BOUND_RECT,
    MP_ROT_BOUND_RECT,
    MP_ENCLOSING_CIRCLE,
]

colors_dict = dict(
    black=C_BLACK,
    blue=C_BLUE,
    blue_violet=C_BLUE_VIOLET,
    blue_cabin=C_CABIN_BLUE,
    cyan=C_CYAN,
    dim_grey=C_DIM_GRAY,
    fuchsia=C_FUCHSIA,
    green=C_GREEN,
    light_steel_blue=C_LIGHT_STEEL_BLUE,
    lime=C_LIME,
    maroon=C_MAROON,
    orange=C_ORANGE,
    purple=C_PURPLE,
    red=C_RED,
    silver=C_SILVER,
    teal=C_TEAL,
    white=C_WHITE,
    yellow=C_YELLOW,
    brown=C_BROWN,
    sienna=C_SIENNA,
)

COLORS = [v for v in colors_dict.values()]

channel_names = ["red", "green", "blue", "hue", "sat", "light"]
channel_colors = {
    "red": "red",
    "green": "green",
    "blue": "blue",
    "hue": "orange",
    "sat": "pink",
    "light": "grey",
}

image_names = {
    DI_NONE: "noImage",
    DI_SEG_SEED_MASK: "segmentationSeedMask",
    DI_SEG_SEED_MASK_KEEP: "segmentationMeedMaskKeep",
    DI_SEG_SEED_MASK_DISCARD: "segmentationSeedMaskDiscard",
    DI_SEG_DISTANCE_MAP: "segmentationDistanceMap",
    DI_SEG_DISTANCE_MAP_KEEP: "segmentationDistanceMapKeep",
    DI_SEG_DISTANCE_MAP_DISCARD: "segmentationDistanceMapDiscard",
    DI_SEG_DISTANCE_MAP_KEEP_WITH_SEEDS: "segmentationDistanceMapKeepWithSeeds",
    DI_SEG_DISTANCE_MAP_DISCARD_WITH_SEEDS: "segmentationDistanceMapDiscardWithSeeds",
    DI_SEG_SEEDS: "segmentationSeeds",
    DI_SEG_RAW_MASK: "segmentationRawMask",
    DI_SEG_LOW_RES_LOGITS_MASK: "segmentationLowResolutionLogitsMask",
    DI_SEG_EVOLVING_MASK: "segmentationMaskProgression",
    DI_SEG_CLEAN_MASK: "segmentationCleanMask",
    DI_SEG_CLEAN_MASK_DEMO: "segmentationCleanMaskDemo",
    DI_SEG_CLEAN_MASK_ON_IMAGE: "segmentationCleanMaskOnImage",
    DI_SEG_CLEAN_MASK_ON_IMAGE_WITH_CONTOURS: "cleanMaskOnImageWithContours",
    DI_SEG_MASK_ON_IMAGE_SUSPECT_CONTOURS: "maskedImage",
    DI_SEG_CLEAN_MASK_ON_IMAGE_WTH_BOXES: "maskedImageWith boxes",
    DI_SEG_CLEAN_MASK_ON_IMAGE_WITH_SEEDS: "segmentationCleanMaskOnImageWithSeeds",
    DI_SEG_PLANT_ONLY: "segmentationPlantOnly",
    DI_SEQ_SRC_IMAGE: "sourceImage",
    DI_SEG_RM_MASK: "rawMask",
    DI_SEQ_RM_PLANT: "plantOnly",
    DI_SEQ_RM_PLANT_HIGHLIGHT: "plantHighlight",
    DF_SEQ_RM_CONTOURS: "plantWithContours",
    DI_SEQ_RM_CONTOURS_HIERARCHY: "plantWithContourHierarchy",
    DI_SEQ_RM_SUSPECT_CONTOURS: "plantAndSuspectContours",
    DI_SEQ_SM_MASK: "siftedMask",
    DI_SEQ_SM_PLANT: "siftedPlant",
    DI_SEQ_SM_CONTOURS: "siftedPlantWithContours",
    DI_SEQ_SM_CONTOURS_ON_SRC: "sourceImageWithSiftedContours",
    DI_SYMP_SEED_MASK: "symptomsSeedMask",
    DI_SYMP_SEED_MASK_KEEP: "symptomsSeedMaskKeep",
    DI_SYMP_SEED_MASK_DISCARD: "symptomsSeedMaskDiscard",
    DI_SYMP_SEEDS: "symptomsSeeds",
}


OPENCV_DIST_METHODS = {
    "Correlation": cv2.HISTCMP_CORREL,
    "Chi-Squared": cv2.HISTCMP_CHISQR,
    "Intersection": cv2.HISTCMP_INTERSECT,
    "Hellinger": cv2.HISTCMP_BHATTACHARYYA,
}

THS_SOURCE = "source"
THS_RAW_MASK_PLANT = "raw plant mask"
THS_MASK_PLANT = "plant mask"
THS_RAW_MASK_BACKGROUND = "raw background mask"
THS_MASK_BACKGROUND = "background mask"
THS_MASK_PLANT_REGIONS = "plant mask regions"
THS_MASK_BACKGROUND_REGIONS = "background mask regions"
THS_MASK_MERGED = "merged mask"
THS_MASK_REGIONS = "regions"
THS_MASK_SEEDS = "merged mask with seeds"
THS_MASK_REGIONS_SEEDS = "regions with seeds"
THS_SEG_MASK_RAW = "raw mask"
THS_SEG_MASK_CLEAN = "clean mask"
THS_SEG_MASK_DEMO = "demo klc mask"
THS_SEG_MASK_LOGITS = "logits"
THS_SEG_MASK_EVO = "segmentation evolution"
THS_SEG_PLANT_ONLY = "plant only"
THS_SEG_PLANT_DEMO = "plant demo"

AVAILABLE_IMAGES = [
    THS_SOURCE,
    THS_RAW_MASK_PLANT,
    THS_MASK_PLANT,
    THS_RAW_MASK_BACKGROUND,
    THS_MASK_BACKGROUND,
    THS_MASK_MERGED,
    THS_MASK_REGIONS,
    THS_MASK_REGIONS_SEEDS,
    THS_MASK_SEEDS,
    THS_SEG_MASK_RAW,
    THS_SEG_MASK_CLEAN,
    THS_SEG_MASK_DEMO,
    THS_SEG_MASK_LOGITS,
    THS_SEG_MASK_EVO,
    THS_SEG_PLANT_ONLY,
    THS_SEG_PLANT_DEMO,
]