Skip to content

Latest commit

 

History

History
52 lines (42 loc) · 1.47 KB

example.md

File metadata and controls

52 lines (42 loc) · 1.47 KB

使い方

DermAnnotationで作成されたTIFF画像を開き,各フレームのラベルを取得

from derm_tiff import load_image

derm_image = load_image("example/imgs/input.tiff")
labels = derm_image.labels
print(labels)  # ['Pencil1', 'Pencil2', 'Pencil3']

DermTiffImageの新規作成

assert isinstance(bg_image, NDArray[Shape["*, *, 3"], np.uint8])
H, W, C = bg_image.shape

empty_derm_image = DermTiffImage(bg_image=bg_image)
new_derm_image = DermTiffImage(bg_image=bg_image,
                               label2mask={"page1": np.zeros((H, W),
                                                              np.bool_)},
                               label2color={"page1": (255, 0, 255)})

TIFF画像の各フレームのマスクを合成して出力

for label in labels:
    pil_img = derm_image.get_annotation_image([label], alpha=0.5)
    pil_img.show()

Pencil1Pencil2Pencil3

新たにフレームを追加

import numpy as np

H, W, _ = derm_image.shape
new_label = "new label"
new_mask = np.random.randint(0, 2, (H, W), dtype=np.bool_)
new_color = (255, 255, 0)
derm_image.add_frame(new_label, new_mask, new_color)

DermAnnotationで開けるTIFF形式で保存

derm_image.save("example/imgs/output.tiff")

全体でリサイズも可能

resized = derm_image.resize(300, 200)