30 napari draw mask

docs/changelog.md

@@ -16,12 +16,16 @@ All notable changes to this project will be documented below.
 
 #### annotate.napari_label_classes
 
-* v0.1dev: viewer = **annotate.napari_label_classes**(*img, classes, size=10, shape='square', importdata=False, show=True*)
+* v0.1dev: viewer = **annotate.napari_label_classes**(*img, classes, size=10, importdata=False, show=True*)
 
 #### annotate.napari_open
 
 * v0.1dev: viewer = **annotate.napari_open**(*img, mode = 'native', show=True*)
 
+#### annotate.napari_points_mask
+
+* v0.1dev: mask_dict = **annotate.napari_points_mask**(*img, viewer*)
+
 #### annotate.napari_read_coor
 
 * v0.1dev: data = **annotate.napari_read_coor**(*coor, dataformat='yx'*)

docs/napari_join_labels.md

@@ -2,17 +2,18 @@
 
 This function joins classes with the same label. This function would be run after classes are labeled with napari_label_classes.
 
-**plantcv.annotate.napari_join_labels*(*img, viewer*)
+**plantcv.annotate.napari_join_labels**(*img, viewer*)
 
 **returns** relabeled mask, dictionary of masks for each class
 
 - **Parameters:**
     - img - image data (compatible with gray, RGB, and hyperspectral data. If data is hyperspecral it should be the array e.g. hyperspectral.array_data)
-    - viewer - viewer with labeled classes. If no points are selected for a class,
+    - viewer - viewer with labeled classes(likely created with [`napari_label_classes`](napari_label_classes.md)).
+        If no points are selected for a class,
         data without labels will default to this class when napari_join_labels
         is run. If all classes have points labeled, any clusters not labeled
         will default to the last class in the list if napari_join_labels is
-        run.
+        run. 
 
 - **Context:**
     - This function would be run after labeling classes in Napari is complete. 

docs/napari_label_classes.md

@@ -3,16 +3,15 @@
 This function opens an image in Napari and then defines a set of Points layers with the user-defined labels called `classes`. A random `shape` of the annotation symbol is assigned to each of the `classes`. 
 Image can be annotated as long as viewer is open. 
 
-**plantcv.annotate.napari_label_classes*(*img, classes, size=10, shape='square', importdata=False, show=True*)
+
+**plantcv.annotate.napari_label_classes**(*img, classes, size=10, importdata=False, show=True*)
 
 **returns** napari viewer object
 
 - **Parameters:**
     - img - image data (compatible with gray, RGB, and hyperspectral data. If data is hyperspecral it should be the array e.g. hyperspectral.array_data)
     - classes - list of classes to label. This option is not necessary if data is data is imported.
     - size - integer pixel size of label (also adjustable from the interactive Napari viewer)
-    - shape - shape of the annotation symbol. Can be 'o', 'arrow', 'clobber', 'cross', 'diamond', 'disc', 'hbar', 'ring', 'square' (default), 'star', 'tailed_arrow', 
-    'triangle_down', 'triangle_up', 'vbar', or 'x' (also adjustable from the interactive Napari viewer)
     - importdata - dictionary of data, data saved from napari_save_coor or data imported from napari_read_coor
     - show - if `show=True`, viewer is launched. `False` setting is useful for test purposes.
 

docs/napari_points_mask.md

@@ -0,0 +1,60 @@
+## Make Mask of Napari Points
+
+This function is to generate a mask from Napari point information.
+This application of this function could be to get information about image
+at particular points (e.g. color or intensity information)
+
+**plantcv.annotate.napari_points_mask**(*img, viewer*)
+
+**returns** dictionary of masks (one for each class where the class label is the key to access)
+
+- **Parameters:**
+    - img - image data (compatible with gray, RGB, and hyperspectral data. If data is hyperspecral it should be the array e.g. hyperspectral.array_data)
+    - viewer = Napari Viewer with point classes labeled (likely created with [`napari_label_classes`](napari_label_classes.md)). The size of the points in the mask will be determined from the viewer parameters.
+
+- **Context:**
+    - This function can be used to generate a mask from Napari points in order to get information about point data. 
+
+- **Example use:**
+    - An application of this function might be collection of color data for the [Naive Bayes module](https://plantcv.readthedocs.io/en/latest/tutorials/machine_learning_tutorial/).
+
+
+```python
+import plantcv.plantcv as pcv 
+import plantcv.annotate as pcvan
+import napari
+
+# Create an instance of the Points class
+img, path, name = pcv.readimage("./wheat.png")
+
+# Should open interactive napari viewer
+viewer = pcvan.napari_label_classes(img=img, classes=['background','healthy', 'rust', 'chlorosis'], size=4)
+
+maskdict = pcvan.napari_points_mask(img, viewer)
+
+pcv.plot_image(maskdict['background'])
+pcv.plot_image(maskdict['healthy'])
+pcv.plot_image(maskdict['rust'])
+pcv.plot_image(maskdict['chlorosis'])
+
+```
+
+![Screenshot](img/documentation_images/napari_points_mask/viewer_labeled.png)
+
+***Background Mask***
+
+![Screenshot](img/documentation_images/napari_points_mask/background.png)
+
+***Healthy Mask***
+
+![Screenshot](img/documentation_images/napari_points_mask/healthy.png)
+
+***Rust Mask***
+
+![Screenshot](img/documentation_images/napari_points_mask/rust.png)
+
+***Chlorosis Mask***
+
+![Screenshot](img/documentation_images/napari_points_mask/chlorosis.png)
+
+**Source Code:** [Here](https://github.com/danforthcenter/plantcv-annotate/blob/main/plantcv/annotate/napari_points_mask.py)

mkdocs.yml

@@ -21,6 +21,7 @@ nav:
       - Napari Join: napari_join_labels.md
       - Napari Label: napari_label_classes.md
       - Napari Open: napari_open.md
+      - Napari Points Mask: napari_points_mask.md
       - Napari Read Coor: napari_read_coor.md
       - Napari Save Coor: napari_save_coor.md
       - Points: Points.md

plantcv/annotate/__init__.py

@@ -7,6 +7,7 @@
 from plantcv.annotate.napari_join_labels import napari_join_labels
 from plantcv.annotate.napari_save_coor import napari_save_coor
 from plantcv.annotate.napari_read_coor import napari_read_coor
+from plantcv.annotate.napari_points_mask import napari_points_mask
 
 # Auto versioning
 __version__ = version("plantcv-annotate")
@@ -19,5 +20,6 @@
     "napari_label_classes",
     "napari_join_labels",
     "napari_save_coor",
-    "napari_read_coor"
+    "napari_read_coor",
+    "napari_points_mask"
 ]

plantcv/annotate/napari_label_classes.py

@@ -6,7 +6,7 @@
 from plantcv.annotate import napari_classes
 
 
-def napari_label_classes(img, classes=False, size=10, shape='square',
+def napari_label_classes(img, classes=False, size=10,
                          importdata=False, show=True):
     """
     open img in napari and label classes
@@ -22,7 +22,6 @@ def napari_label_classes(img, classes=False, size=10, shape='square',
         will default to the last class in the list when napari_join_labels is
         run.
     size = size of marker in pixels
-    shape = either 'square' or 'circle'
     importdata = dictionary of values in Napari format (y,x).
         Output of napari_read_coor
     show = if show is True the viewer is launched. This opetion is useful for
@@ -35,7 +34,6 @@ def napari_label_classes(img, classes=False, size=10, shape='square',
     :param img: numpy.ndarray
     :param classes: list
     :param size: int
-    :param shape: str
     :param importdata: dict
     :param show: str
 
@@ -83,7 +81,7 @@ def napari_label_classes(img, classes=False, size=10, shape='square',
     if classes is not False:
         for x in classes:
             if x not in keys:
-                viewer.add_points(np.array([]), name=x, symbol=shape,
+                viewer.add_points(np.array([]), name=x, symbol='square',
                                   edge_color=random.choice(color),
                                   face_color=random.choice(color), size=size)
         keys = napari_classes(viewer)
@@ -95,7 +93,8 @@ def napari_label_classes(img, classes=False, size=10, shape='square',
                 if key in keys:
                     viewer.layers[key].add(importdata[key])
                 else:
-                    viewer.add_points(importdata[key], name=key, symbol=shape,
+                    viewer.add_points(importdata[key], name=key,
+                                      symbol='square',
                                       edge_color=random.choice(color),
                                       face_color=random.choice(color),
                                       size=size)

plantcv/annotate/napari_open.py

@@ -2,8 +2,8 @@
 
 import cv2
 import numpy as np
-from skimage.color import label2rgb
 import napari
+from skimage.color import label2rgb
 
 
 def napari_open(img, mode='native', show=True):

plantcv/annotate/napari_points_mask.py

@@ -0,0 +1,55 @@
+# Make Masks of Labelled Napari Points
+
+import numpy as np
+import os
+import cv2
+from plantcv.annotate import napari_classes
+from plantcv.plantcv import params
+from plantcv.plantcv._debug import _debug
+
+
+def napari_points_mask(img, viewer):
+    """
+    draw points mask based on Napari viewer annotations
+
+    Inputs:
+    img  = img  (grayimg, rgbimg, or hyperspectral image array data
+        e.g. hyperspectraldata.array_data). This is used to find the x,y size
+        of the image
+    viewer = Napari Viewer with classes labeled. The size of the masked points
+        will be from the viewer parameters
+
+    Returns:
+    mask_dict   = dictionary of masks; mask for each labelled class
+
+    :param img: numpy.ndarray
+    :param viewer: Napari Viewer object
+    :param shape: str
+    :return mask_dict: dict of numpy.ndarray
+
+    """
+    # get shape of image
+    size = np.shape(img)
+    keys = napari_classes(viewer)
+    maskdict = {}
+
+    for key in keys:
+        maskname = str(key)
+        mask = np.zeros((size[0], size[1]))
+        data = list(viewer.layers[key].data)
+        shapesize = viewer.layers[key]._current_size
+        shapesizehalf = int(shapesize/2)
+
+        for y, x in data:
+            startpoint = (int(x-shapesizehalf), int(y-shapesizehalf))
+            endpoint = (int(x+shapesizehalf-1), int(y+shapesizehalf-1))
+            mask = cv2.rectangle(mask, startpoint, endpoint, (255), -1)
+
+        maskdict[maskname] = mask
+
+        _debug(visual=mask, filename=os.path.join(params.debug_outdir,
+                                                  str(params.device) +
+                                                  str(maskname) +
+                                                  '_labeled_mask.png'))
+
+    return maskdict

tests/test_napari_classes.py

@@ -9,9 +9,9 @@ def test_napari_classes(make_napari_viewer):
     img = np.zeros((100, 100))
     coor = [(25, 25), (50, 50)]
     viewer.add_image(img)
-    viewer.add_points(np.array(coor), symbol="o", name="total",
+    viewer.add_points(np.array(coor), symbol="square", name="total",
                       face_color="red", size=30)
-    viewer.add_points(np.array(coor), symbol="o", name="test",
+    viewer.add_points(np.array(coor), symbol="square", name="test",
                       face_color="red", size=30)
     keys = napari_classes(viewer)
 

tests/test_napari_join_labels.py

@@ -1,8 +1,8 @@
 import numpy as np
-from plantcv.annotate.napari_open import napari_open
-from plantcv.annotate.napari_label_classes import napari_label_classes
+from plantcv.annotate import napari_open
+from plantcv.annotate import napari_label_classes
 from plantcv.plantcv import readimage
-from plantcv.annotate.napari_join_labels import napari_join_labels
+from plantcv.annotate import napari_join_labels
 from plantcv.plantcv import params
 
 
@@ -23,10 +23,10 @@ def test_napari_join_allclass(test_data):
     img, _, _ = readimage(test_data.kmeans_seed_gray_img)
     viewer = napari_open(img, show=False)
     background = [(54, 143), (77, 246)]
-    viewer.add_points(np.array(background), symbol="o", name='background',
+    viewer.add_points(np.array(background), symbol="square", name='background',
                       face_color="red", size=1)
     wing = [(275, 54)]
-    viewer.add_points(np.array(wing), symbol="o", name='wing',
+    viewer.add_points(np.array(wing), symbol="square", name='wing',
                       face_color="red", size=1)
     seed = [(280, 218)]
     viewer.add_points(np.array(seed), symbol="o", name='seed',
@@ -44,13 +44,13 @@ def test_napari_join_warn(test_data):
     img, _, _ = readimage(test_data.kmeans_seed_gray_img)
     viewer = napari_open(img, show=False)
     background = [(54, 143), (77, 246)]
-    viewer.add_points(np.array(background), symbol="o", name='background',
+    viewer.add_points(np.array(background), symbol="square", name='background',
                       face_color="red", size=1)
     wing = [(275, 54)]
-    viewer.add_points(np.array(wing), symbol="o", name='wing',
+    viewer.add_points(np.array(wing), symbol="square", name='wing',
                       face_color="red", size=1)
     seed = [(275, 54)]
-    viewer.add_points(np.array(seed), symbol="o", name='seed',
+    viewer.add_points(np.array(seed), symbol="square", name='seed',
                       face_color="red", size=1)
 
     labeled, _ = napari_join_labels(img, viewer)
@@ -68,13 +68,13 @@ def test_napari_join_print(test_data, tmpdir):
     img, _, _ = readimage(test_data.kmeans_seed_gray_img)
     viewer = napari_open(img, show=False)
     background = [(54, 143), (77, 246)]
-    viewer.add_points(np.array(background), symbol="o", name='background',
+    viewer.add_points(np.array(background), symbol="square", name='background',
                       face_color="red", size=1)
     wing = [(280, 218)]
-    viewer.add_points(np.array(wing), symbol="o", name='wing',
+    viewer.add_points(np.array(wing), symbol="square", name='wing',
                       face_color="red", size=1)
     seed = [(275, 54)]
-    viewer.add_points(np.array(seed), symbol="o", name='seed',
+    viewer.add_points(np.array(seed), symbol="square", name='seed',
                       face_color="red", size=1)
 
     labeled, _ = napari_join_labels(img, viewer)

tests/test_napari_label_classes.py

@@ -11,7 +11,7 @@ def test_napari_label_classes_gray(test_data):
     viewer = napari_label_classes(img, ['total'], size=5, importdata=data,
                                   show=False)
     coor = [(50, 25)]
-    viewer.add_points(np.array(coor), symbol="o", name='coor',
+    viewer.add_points(np.array(coor), symbol="square", name='coor',
                       face_color="red", size=5)
 
     assert len(viewer.layers['total'].data) == 1

tests/test_napari_open.py

@@ -22,7 +22,7 @@ def test_napari_open_gray(test_data):
     img, _, _ = readimage(test_data.kmeans_seed_gray_img)
     viewer = napari_open(img, show=False)
     coor = [(25, 25), (50, 50)]
-    viewer.add_points(np.array(coor), symbol="o", name="total",
+    viewer.add_points(np.array(coor), symbol="square", name="total",
                       face_color="red", size=1)
 
     assert len(viewer.layers['total'].data) == 2
@@ -49,7 +49,7 @@ def test_napari_open_envi(test_data):
     img = img.array_data
     viewer = napari_open(img, show=False)
     coor = [(25, 25), (50, 50)]
-    viewer.add_points(np.array(coor), symbol="o", name="total",
+    viewer.add_points(np.array(coor), symbol="square", name="total",
                       face_color="red", size=1)
 
     assert len(viewer.layers['total'].data) == 2

tests/test_napari_points_mask.py

@@ -0,0 +1,19 @@
+from plantcv.annotate import napari_points_mask
+from plantcv.plantcv import readimage
+from plantcv.annotate import napari_label_classes
+import numpy as np
+
+
+def test_napari_points_mask(test_data):
+    """Test for PlantCV.Annotate"""
+    # Read in test data
+    img, _, _ = readimage(test_data.small_rgb_img)
+    data = {'total': [(25, 25)], 'background': [(50, 50)]}
+    viewer = napari_label_classes(img, ['total'], size=50, importdata=data,
+                                  show=False)
+    maskdict = napari_points_mask(img, viewer)
+
+    summask = int((np.sum(maskdict['total']))/255)
+
+    assert summask == 2500
+    viewer.close()

tests/test_napari_save_coor.py

@@ -12,7 +12,7 @@ def test_napari_save_coor(test_data, tmpdir):
     img, _, _ = readimage(test_data.kmeans_seed_gray_img)
     viewer = napari_label_classes(img, ['seed'], show=False)
     coor = [(25, 25)]
-    viewer.add_points(np.array(coor), symbol="o", name='background',
+    viewer.add_points(np.array(coor), symbol="square", name='background',
                       face_color="red", size=1)
 
     filename = os.path.join(cache_dir, 'tempfile.txt')