Vizailization will be created here

Viz notebook

The autoreload extension is already loaded. To reload it, use:
  %reload_ext autoreload

mpl.rcParams['image.cmap'] = 'gray'

Path.home()

Path('/home/hasan')

im_path = Path(r'/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/')
mask_path = Path(r'/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_masks/')
#im_path = Path(r'/home/hasan/workspace/data/microscopy_data/patch_images/')
#mask_path = Path(r'/home/hasan/workspace/data/microscopy_data/patch_masks')

images=im_path.ls()
masks=mask_path.ls()

r_idx = np.random.choice(len(im_path.ls()), 1)[0]

s_img = images[r_idx]
s_msk = masks[r_idx]

s_img.name, s_msk.name

('img_21_p_3.png', 'img_21_p_3.png')

sn_msk_img = cv2.imread(str(s_msk), cv2.IMREAD_GRAYSCALE)
np.unique(sn_msk_img)

array([  0, 255], dtype=uint8)

source

overlay_mask

 overlay_mask (im_path, msk_path, overlay_clr=(0, 1, 0), scale=1,
               alpha=0.5)

Creaete a overlay image from image and mask

overlay_mask(s_img, s_msk,alpha=0.1)

source

overlay_mask_border_on_image

 overlay_mask_border_on_image (im_path, msk_path, border_color=(0, 1, 0),
                               border_width=1)

*Overlays the border of a binary mask on a grayscale image and displays the result using matplotlib.

Args: image (numpy.ndarray): Grayscale image. mask (numpy.ndarray): Binary mask of the same size as the image. border_color (tuple): RGB color for the mask border in the range [0, 1]. border_width (int): Width of the border.

Returns: None: The function displays a plot.*

overlay_mask_border_on_image(s_img, s_msk)

#yolo_mask_path = Path(r'/home/hasan/workspace/data/microscopy_data/yolo_dataset_train')
#im_path = Path(r'/home/hasan/workspace/data/microscopy_data/patch_images/')
yolo_mask_path = Path(r'/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train')
im_path = Path(r'/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/')
yolo_mask_path.ls(), im_path.ls()

((#1633) [Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_139_p_11.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_48_p_0.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_120_p_1.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_54_p_9.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_2_p_2.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_34_p_10.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_50_p_1.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_42_p_8.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_67_p_1.txt'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_20_p_1.txt')...],
 (#1642) [Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_162_p_9.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_11_p_9.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_70_p_5.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_67_p_1.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_74_p_9.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_16_p_5.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_23_p_1.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_112_p_6.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_121_p_11.png'),Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_152_p_1.png')...])

r_idx = np.random.choice(len(im_path.ls()), 1)[0]
print(f'random index: {r_idx}')

random index: 546

yolo_mask_path.ls()[r_idx], im_path.ls()[r_idx]

(Path('/home/hasan/Schreibtisch/projects/data/microscopy/yolo_dataset_train/img_27_p_0.txt'),
 Path('/home/hasan/Schreibtisch/projects/data/microscopy/patch_train_images/img_73_p_6.png'))

source

overlay_yolo_mask

 overlay_yolo_mask (im_path, msk_path, color=None, idx=None, im_name=None,
                    fill=False, alpha=0.3)

Creaete a overlay image from image and mask

	Type	Default	Details
im_path
msk_path
color	NoneType	None	color values other matplotlib colors
idx	NoneType	None	Index of image in the folder other random will be chosen
im_name	NoneType	None	Imge name with extension
fill	bool	False	whther to fill the mask or not
alpha	float	0.3	alpha value for the mask for transparency

overlay_yolo_mask(im_path, msk_path=yolo_mask_path, color='red', fill=False)