Merge branch 'main' into napari-join-labels

.github/workflows/continuous-integration.yml

@@ -16,7 +16,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        python-version: ['3.8', '3.9', '3.10']
+        python-version: ['3.9', '3.10', '3.11']
         os: [ubuntu-latest]
     env:
       OS: ${{ matrix.os }}

docs/get_centroids.md

@@ -0,0 +1,44 @@
+## Get Centroids 
+
+Extract the centroid coordinate (column,row) from regions in a binary image. 
+
+**plantcv.annotate.get_centroids**(*bin_img*)
+
+**returns** list containing coordinates of centroids 
+
+- **Parameters:**
+    - bin_img - Binary image containing the connected regions to consider
+- **Context:**
+    - Given an arbitrary mask of the objects of interest, `get_centroids`
+    returns a list of coordinates that can the be imported into the annotation class [Points](Points.md).
+
+- **Example use:**
+    - Below
+
+**Binary image**
+
+![count_img](img/documentation_images/get_centroids/discs_mask.png)
+
+```python
+
+from plantcv import plantcv as pcv
+
+# Set global debug behavior to None (default), "print" (to file),
+# or "plot"
+pcv.params.debug = "plot"
+
+# Apply get centroids to the binary image
+coords = pcv.annotate.get_centroids(bin_img=binary_img)
+print(coords)
+# [[1902, 600], [1839, 1363], [1837, 383], [1669, 1977], [1631, 1889], [1590, 1372], [1550, 1525],
+# [1538, 1633], [1522, 1131], [1494, 2396], [1482, 1917], [1446, 1808], [1425, 726], [1418, 2392],
+# [1389, 198], [1358, 1712], [1288, 522], [1289, 406], [1279, 368], [1262, 1376], [1244, 1795],
+# [1224, 1327], [1201, 624], [1181, 725], [1062, 85], [999, 840], [885, 399], [740, 324], [728, 224],
+# [697, 860], [660, 650], [638, 2390], [622, 1565], [577, 497], [572, 2179], [550, 2230], [547, 1826],
+# [537, 892], [538, 481], [524, 2144], [521, 2336], [497, 201], [385, 1141], [342, 683], [342, 102],
+# [332, 1700], [295, 646], [271, 60], [269, 1626], [210, 1694], [189, 878], [178, 1570], [171, 2307],
+# [61, 286], [28, 2342]]
+
+```
+
+**Source Code:** [Here](https://github.com/danforthcenter/plantcv-annotate/blob/main/plantcv/annotate/get_centroids.py)

mkdocs.yml

@@ -22,7 +22,7 @@ nav:
       - Napari Label: napari_label_classes.md
       - Napari Open: napari_open.md
       - Points: Points.md
-
+      - Get Centroids: get_centroids.md
 markdown_extensions:
   - toc:
       permalink: True

plantcv/annotate/__init__.py

@@ -1,12 +1,19 @@
+from importlib.metadata import version
+from plantcv.annotate.classes import Points
+from plantcv.annotate.get_centroids import get_centroids
 from plantcv.annotate.napari_classes import napari_classes
 from plantcv.annotate.napari_open import napari_open
 from plantcv.annotate.napari_label_classes import napari_label_classes
 from plantcv.annotate.napari_join_labels import napari_join_labels
 
+# Auto versioning
+__version__ = version("plantcv-annotate")
 
 __all__ = [
+    "Points",
+    "get_centroids",
     "napari_classes",
     "napari_open",
     "napari_label_classes",
     "napari_join_labels"
-]
+    ]

plantcv/annotate/classes.py

@@ -0,0 +1,160 @@
+# Class helpers
+
+# Imports
+import cv2
+import json
+from math import floor
+import matplotlib.pyplot as plt
+from plantcv.plantcv.annotate.points import _find_closest_pt
+from plantcv.plantcv import warn
+
+
+class Points:
+    """Point annotation/collection class to use in Jupyter notebooks. It allows the user to
+    interactively click to collect coordinates from an image. Left click collects the point and
+    right click removes the closest collected point.
+    """
+
+    def __init__(self, img, figsize=(12, 6), label="default", color="r", view_all=False):
+        """Points initialization method.
+
+        Parameters
+        ----------
+        img : numpy.ndarray
+            image to annotate
+        figsize : tuple, optional
+            figure plotting size, by default (12, 6)
+        label : str, optional
+            class label, by default "default"
+        """
+        self.img = img
+        self.figsize = figsize
+        self.label = label  # current label
+        self.color = color  # current color
+        self.view_all = view_all  # a flag indicating whether or not view all labels
+        self.coords = {}  # dictionary of all coordinates per group label
+        self.events = []  # includes right and left click events
+        self.count = {}  # a dictionary that saves the counts of different groups (labels)
+        self.sample_labels = []  # list of all sample labels, one to one with points collected
+        self.colors = {}  # all used colors
+
+        self.view(label=self.label, color=self.color, view_all=self.view_all)
+
+    def onclick(self, event):
+        """Handle mouse click events
+
+        Parameters
+        ----------
+        event : matplotlib.backend_bases.MouseEvent
+            matplotlib MouseEvent object
+        """
+        print(type(event))
+        self.events.append(event)
+        if event.button == 1:
+            # Add point to the plot
+            self.ax.plot(event.xdata, event.ydata, marker='x', c=self.color)
+            self.coords[self.label].append((floor(event.xdata), floor(event.ydata)))
+            self.count[self.label] += 1
+            self.sample_labels.append(self.label)
+        else:
+            idx_remove, _ = _find_closest_pt((event.xdata, event.ydata), self.coords[self.label])
+            # remove the closest point to the user right clicked one
+            self.coords[self.label].pop(idx_remove)
+            self.count[self.label] -= 1
+            idx_remove = idx_remove + self.p_not_current
+            self.ax.lines[idx_remove].remove()
+            self.sample_labels.pop(idx_remove)
+        self.fig.canvas.draw()
+
+    def print_coords(self, filename):
+        """Save collected coordinates to a file.
+
+        Parameters
+        ----------
+        filename : str
+            output filename
+        """
+        # Open the file for writing
+        with open(filename, "w") as fp:
+            # Save the data in JSON format with indentation
+            json.dump(obj=self.coords, fp=fp, indent=4)
+
+    def import_list(self, coords, label="default"):
+        """Import coordinates.
+
+        Parameters
+        ----------
+        coords : list
+            list of coordinates (tuples)
+        label : str, optional
+            class label, by default "default"
+        """
+        if label not in self.coords:
+            self.coords[label] = []
+            for (x, y) in coords:
+                self.coords[label].append((x, y))
+            self.count[label] = len(self.coords[label])
+            self.view(label=label, color=self.color, view_all=False)
+        else:
+            warn(f"{label} already included and counted, nothing is imported!")
+
+    def import_file(self, filename):
+        """Import coordinates from a file.
+
+        Parameters
+        ----------
+        filename : str
+            JSON file containing Points annotations
+        """
+        with open(filename, "r") as fp:
+            coords = json.load(fp)
+
+        keys = list(coords.keys())
+
+        for key in keys:
+            keycoor = coords[key]
+            keycoor = list(map(lambda sub: (sub[1], sub[0]), keycoor))
+            self.import_list(keycoor, label=key)
+
+    def view(self, label="default", color="r", view_all=False):
+        """View coordinates for a specific class label.
+
+        Parameters
+        ----------
+        label : str, optional
+            class label, by default "default"
+        color : str, optional
+            marker color, by default "r"
+        view_all : bool, optional
+            view all classes or a single class, by default False
+        """
+        if label not in self.coords and color in self.colors.values():
+            warn("The color assigned to the new class label is already used, if proceeding, "
+                 "items from different classes will not be distinguishable in plots!")
+        self.label = label
+        self.color = color
+        self.view_all = view_all
+
+        if self.label not in self.coords:
+            self.coords[self.label] = []
+            self.count[self.label] = 0
+        self.colors[self.label] = self.color
+
+        self.fig, self.ax = plt.subplots(1, 1, figsize=self.figsize)
+
+        self.events = []
+        self.fig.canvas.mpl_connect('button_press_event', self.onclick)
+
+        self.ax.imshow(cv2.cvtColor(self.img, cv2.COLOR_BGR2RGB))
+        self.ax.set_title("Please left click on objects\n Right click to remove")
+        self.p_not_current = 0
+        # if view_all is True, show all already marked markers
+        if self.view_all:
+            for k in self.coords:
+                for (x, y) in self.coords[k]:
+                    self.ax.plot(x, y, marker='x', c=self.colors[k])
+                    if self.label not in self.coords or len(self.coords[self.label]) == 0:
+                        self.p_not_current += 1
+        else:
+            for (x, y) in self.coords[self.label]:
+                self.ax.plot(x, y, marker='x', c=self.color)

plantcv/annotate/get_centroids.py

@@ -0,0 +1,28 @@
+# Get centroids from mask objects
+
+from skimage.measure import label, regionprops
+
+
+def get_centroids(bin_img):
+    """Get the coordinates (row,column) of the centroid of each connected region in a binary image.
+
+    Inputs:
+    bin_img         = Binary image containing the connected regions to consider
+
+    Returns:
+    coords          = List of coordinates (row,column) of the centroids of the regions
+
+    :param bin_img: numpy.ndarray
+    :return coords: list
+    """
+    # find contours in the binary image
+    labeled_img = label(bin_img)
+    # measure regions
+    obj_measures = regionprops(labeled_img)
+    coords = []
+    for obj in obj_measures:
+        # Convert coord values to int
+        coord = tuple(map(int, obj.centroid))
+        coords.append(coord)
+
+    return coords

pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["setuptools >= 61.0"]
+requires = ["setuptools >= 64.0", "setuptools_scm>=8"]
 build-backend = "setuptools.build_meta"
 
 [tool.setuptools.packages.find]
@@ -27,12 +27,17 @@ classifiers = [
     "License :: OSI Approved :: Mozilla Public License 2.0 (MPL 2.0)",
     "Intended Audience :: Science/Research",
 ]
+
 [project.optional-dependencies]
 test = [
     "pytest",
+    "pytest-cov",
     "pytest-qt"
 ]
+
 [project.urls]
 Homepage = "https://plantcv.org"
 Documentation = "https://plantcv.readthedocs.io"
-Repository = "https://github.com/danforthcenter/plantcv-annotate"
+Repository = "https://github.com/danforthcenter/plantcv-annotate"
+
+[tool.setuptools_scm]

tests/conftest.py

@@ -1,6 +1,6 @@
+import os
 import pytest
 import matplotlib
-import os
 
 
 # Disable plotting
@@ -11,17 +11,19 @@ class TestData:
     def __init__(self):
         """Initialize simple variables."""
         # Test data directory
-        self.datadir = os.path.join(os.path.dirname(os.path.abspath(__file__)),
-                                    "testdata")
+        self.datadir = os.path.join(os.path.dirname(os.path.abspath(__file__)), "testdata")
+        # Flat image directory
+        self.snapshot_dir = os.path.join(self.datadir, "snapshot_dir")
         # RGB image
-        filename_rgb = "setaria_small_plant_rgb.png"
-        self.small_rgb_img = os.path.join(self.datadir, filename_rgb)
+        self.small_rgb_img = os.path.join(self.datadir, "setaria_small_plant_rgb.png")
+        # Binary image
+        self.small_bin_img = os.path.join(self.datadir, "setaria_small_plant_mask.png")
+        # Text file with tuple coordinates (by group label)
+        self.pollen_coords = os.path.join(self.datadir, "points_file_import.coords")
         # Kmeans Clustered Gray image
-        filename_kmeans = "silphium_seed_labeled_example.png"
-        self.kmeans_seed_gray_img = os.path.join(self.datadir, filename_kmeans)
+        self.kmeans_seed_gray_img = os.path.join(self.datadir, "silphium_seed_labeled_example.png")
         # Small Hyperspectral image
-        filename_hyper = "corn-kernel-hyperspectral.raw"
-        self.envi_sample_data = os.path.join(self.datadir, filename_hyper)
+        self.envi_sample_data = os.path.join(self.datadir, "corn-kernel-hyperspectral.raw")
 
 
 @pytest.fixture(scope="session")

tests/test_annotate_get_centroids.py

@@ -0,0 +1,12 @@
+import cv2
+from plantcv.annotate.get_centroids import get_centroids
+
+
+def test_get_centroids(test_data):
+    """Test for PlantCV."""
+    # Read in test data
+    mask = cv2.imread(test_data.small_bin_img, -1)
+
+    coor = get_centroids(bin_img=mask)
+
+    assert coor == [(166, 214)]