tau-fibrils-yolo 0.0.6__tar.gz → 0.1.1__tar.gz

{tau_fibrils_yolo-0.0.6 → tau_fibrils_yolo-0.1.1}/.github/workflows/release_multiplatform.yml

@@ -23,7 +23,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: "3.9"
+          python-version: "3.10"
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
@@ -74,7 +74,7 @@ jobs:
       - name: Build executable
         env:
           PYAPP_PROJECT_VERSION: ${{ env.VERSION }}
-          PYAPP_PYTHON_VERSION: '3.9'
+          PYAPP_PYTHON_VERSION: '3.10'
         run: cargo build --release --manifest-path pyapp-latest/Cargo.toml
 
       - name: Archive executable
@@ -96,7 +96,7 @@ jobs:
           fi
 
       - name: Upload artifact
-        uses: actions/upload-artifact@v2
+        uses: actions/upload-artifact@v4
         with:
           name: executable-${{ runner.os }}
           path: executable-*.tar.gz
@@ -109,19 +109,19 @@ jobs:
         uses: actions/checkout@v2
 
       - name: Download Windows artifact
-        uses: actions/download-artifact@v2
+        uses: actions/download-artifact@v4
         with:
           name: executable-Windows
           path: .
 
       - name: Download Linux artifact
-        uses: actions/download-artifact@v2
+        uses: actions/download-artifact@v4
         with:
           name: executable-Linux
           path: .
       
       - name: Download MacOS artifact
-        uses: actions/download-artifact@v2
+        uses: actions/download-artifact@v4
         with:
           name: executable-macOS
           path: .

tau_fibrils_yolo-0.1.1/.gitignore

@@ -0,0 +1,34 @@
+__pycache__/
+*.py[cod]
+*$py.class
+old/
+data/
+env/
+build/
+dist/
+downloads/
+eggs/
+.eggs/
+*.egg-info/
+.installed.cfg
+*.egg
+.tox/
+.coverage
+.coverage.*
+.cache
+docs/_build/
+.idea/
+venv/
+.vscode/
+.ipynb_checkpoints
+.python-version
+.DS_Store
+**/_version.py
+
+push.sh
+todo.txt
+trained_models/
+pretrained_models/
+qupath/
+output/
+notebooks/

{tau_fibrils_yolo-0.0.6 → tau_fibrils_yolo-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: tau-fibrils-yolo
-Version: 0.0.6
+Version: 0.1.1
 Summary: YoloV8 model for the detection of Tau fibrils in EM images.
 Author-email: Mallory Wittwer <mallory.wittwer@epfl.ch>
 License:                     GNU AFFERO GENERAL PUBLIC LICENSE
@@ -672,6 +672,9 @@ Classifier: License :: OSI Approved :: BSD License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Scientific/Engineering :: Image Processing
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
@@ -683,18 +686,20 @@ Requires-Dist: numpy
 Requires-Dist: pandas
 Requires-Dist: tifffile
 Requires-Dist: matplotlib
-Requires-Dist: pooch==1.8.0
 Requires-Dist: scikit-image
 Requires-Dist: scikit-learn
 Requires-Dist: ultralytics
 Requires-Dist: opencv-contrib-python-headless
+Requires-Dist: imaging-server-kit==0.1.3
+Requires-Dist: napari-serverkit==0.0.9
+Dynamic: license-file
 
 ![EPFL Center for Imaging logo](https://imaging.epfl.ch/resources/logo-for-gitlab.svg)
 # 🧬 Tau Fibrils Yolo - Object detection in EM images
 
 ![screenshot](assets/screenshot.png)
 
-We provide a [YoloV8](https://docs.ultralytics.com/) model for the detection of oriented bounding boxes (OBBs) of Tau fibrils in EM images.
+We provide a [YoloV8](https://docs.ultralytics.com/) model for the detection of oriented bounding boxes (OBBs) of Tau fibrils in EM images. The model is integrated as a [Napari](https://napari.org/stable/) plugin.
 
 [[`Installation`](#installation)] [[`Model`](#model)] [[`Usage`](#usage)] [[`Training`](#training)]
 
@@ -702,13 +707,7 @@ This project is part of a collaboration between the [EPFL Center for Imaging](ht
 
 ## Installation
 
-### As a standalone app
-
-Download and run the latest installer from the [Releases](https://github.com/EPFL-Center-for-Imaging/tau-fibrils-yolo/releases) page.
-
-### As a Python package
-
-We recommend performing the installation in a clean Python environment. Install our package from PyPi:
+We recommend performing the installation in a clean Python environment. Install the package from PyPi:
 
 ```sh
 pip install tau-fibrils-yolo
@@ -728,71 +727,32 @@ cd tau-fibrils-yolo
 pip install -e .
 ```
 
-## Model
-
-The model weights (6.5 Mb) are automatically downloaded from [this repository on Zenodo](https://sandbox.zenodo.org/records/99113) the first time you run inference. The model files are saved in the user home folder in the `.yolo` directory.
-
 ## Usage
 
 **In Napari**
 
-To use our model in Napari, start the viewer with
+To use the model in Napari, start the viewer with
 
 ```sh
 napari -w tau-fibrils-yolo
 ```
 
-or open the Napari menu bar and select `Plugins > Tau fibrils detection`.
-
-Open an image using `File > Open files` or drag-and-drop an image into the viewer window.
-
-**Sample data**: To test the model, you can run it on our provided sample image. In Napari, open the image from `File > Open Sample > [TODO - add a sample image]`.
-
-
-**As a library**
-
-You can run the model to detect fibrils in an image (represented as a numpy array).
-
-```py
-from tau_fibrils_yolo import FibrilsDetector
-
-detector = FibrilsDetector()
-
-boxes, probabilities = detector.predict(your_image)
-```
+or open the plugin from `Plugins > Tau fibrils detection`.
 
-**As a CLI**
+**From the command-line**
 
-Run inference on an image from the command-line. For example:
+Run inference on an image from the command-line:
 
 ```sh
 tau_fibrils_predict_image -i /path/to/folder/image_001.tif
 ```
 
-The command will save the segmentation next to the image:
-
-```
-folder/
-    ├── image_001.tif
-    ├── image_001_results.csv
-```
-
-Optionally, you can use the `-r` flag to also rescale the image by a given factor.
-
-To run inference in batch on all images in a folder, use:
-
-```sh
-tau_fibrils_predict_folder -i /path/to/folder/
-```
-
-This will produce:
+This command will run the YOLO model and save a CSV file containing measurements next to the image:
 
 ```
 folder/
     ├── image_001.tif
     ├── image_001_results.csv
-    ├── image_002.tif
-    ├── image_002_results.csv
 ```
 
 ## Training

{tau_fibrils_yolo-0.0.6 → tau_fibrils_yolo-0.1.1}/README.md

@@ -3,7 +3,7 @@
 
 ![screenshot](assets/screenshot.png)
 
-We provide a [YoloV8](https://docs.ultralytics.com/) model for the detection of oriented bounding boxes (OBBs) of Tau fibrils in EM images.
+We provide a [YoloV8](https://docs.ultralytics.com/) model for the detection of oriented bounding boxes (OBBs) of Tau fibrils in EM images. The model is integrated as a [Napari](https://napari.org/stable/) plugin.
 
 [[`Installation`](#installation)] [[`Model`](#model)] [[`Usage`](#usage)] [[`Training`](#training)]
 
@@ -11,13 +11,7 @@ This project is part of a collaboration between the [EPFL Center for Imaging](ht
 
 ## Installation
 
-### As a standalone app
-
-Download and run the latest installer from the [Releases](https://github.com/EPFL-Center-for-Imaging/tau-fibrils-yolo/releases) page.
-
-### As a Python package
-
-We recommend performing the installation in a clean Python environment. Install our package from PyPi:
+We recommend performing the installation in a clean Python environment. Install the package from PyPi:
 
 ```sh
 pip install tau-fibrils-yolo
@@ -37,71 +31,32 @@ cd tau-fibrils-yolo
 pip install -e .
 ```
 
-## Model
-
-The model weights (6.5 Mb) are automatically downloaded from [this repository on Zenodo](https://sandbox.zenodo.org/records/99113) the first time you run inference. The model files are saved in the user home folder in the `.yolo` directory.
-
 ## Usage
 
 **In Napari**
 
-To use our model in Napari, start the viewer with
+To use the model in Napari, start the viewer with
 
 ```sh
 napari -w tau-fibrils-yolo
 ```
 
-or open the Napari menu bar and select `Plugins > Tau fibrils detection`.
-
-Open an image using `File > Open files` or drag-and-drop an image into the viewer window.
-
-**Sample data**: To test the model, you can run it on our provided sample image. In Napari, open the image from `File > Open Sample > [TODO - add a sample image]`.
-
-
-**As a library**
-
-You can run the model to detect fibrils in an image (represented as a numpy array).
-
-```py
-from tau_fibrils_yolo import FibrilsDetector
-
-detector = FibrilsDetector()
-
-boxes, probabilities = detector.predict(your_image)
-```
+or open the plugin from `Plugins > Tau fibrils detection`.
 
-**As a CLI**
+**From the command-line**
 
-Run inference on an image from the command-line. For example:
+Run inference on an image from the command-line:
 
 ```sh
 tau_fibrils_predict_image -i /path/to/folder/image_001.tif
 ```
 
-The command will save the segmentation next to the image:
-
-```
-folder/
-    ├── image_001.tif
-    ├── image_001_results.csv
-```
-
-Optionally, you can use the `-r` flag to also rescale the image by a given factor.
-
-To run inference in batch on all images in a folder, use:
-
-```sh
-tau_fibrils_predict_folder -i /path/to/folder/
-```
-
-This will produce:
+This command will run the YOLO model and save a CSV file containing measurements next to the image:
 
 ```
 folder/
     ├── image_001.tif
     ├── image_001_results.csv
-    ├── image_002.tif
-    ├── image_002_results.csv
 ```
 
 ## Training

{tau_fibrils_yolo-0.0.6 → tau_fibrils_yolo-0.1.1}/pyproject.toml

@@ -13,6 +13,9 @@ classifiers = [
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
     "Topic :: Scientific/Engineering :: Image Processing",
 ]
 
@@ -24,11 +27,12 @@ dependencies = [
     "pandas",
     "tifffile",
     "matplotlib",
-    "pooch==1.8.0",
     "scikit-image",
     "scikit-learn",
     "ultralytics",
-    "opencv-contrib-python-headless"
+    "opencv-contrib-python-headless",
+    "imaging-server-kit==0.1.3",
+    "napari-serverkit==0.0.9",
 ]
 
 [project.entry-points."napari.manifest"]

tau_fibrils_yolo-0.1.1/src/tau_fibrils_yolo/__init__.py

@@ -0,0 +1,28 @@
+from tau_fibrils_yolo.widget import (
+    detect_fibrils,
+    max_iou_boxes,
+    min_distance_boxes,
+    boxes_kernel_density,
+)
+
+from qtpy.QtWidgets import QWidget
+import imaging_server_kit as sk
+
+
+class YoloDetectorWidget(QWidget):
+    def __init__(self, napari_viewer):
+        super().__init__()
+        multi = sk.combine(
+            [detect_fibrils, max_iou_boxes, min_distance_boxes, boxes_kernel_density],
+            name="Tau fibrils analysis",
+        )
+        widget = sk.to_qwidget(multi, napari_viewer)
+        self.setLayout(widget.layout())
+
+
+from ._version import version as __version__
+
+try:
+    from ._version import version as __version__
+except ImportError:
+    __version__ = "unknown"

tau_fibrils_yolo-0.1.1/src/tau_fibrils_yolo/_version.py

@@ -0,0 +1,34 @@
+# file generated by setuptools-scm
+# don't change, don't track in version control
+
+__all__ = [
+    "__version__",
+    "__version_tuple__",
+    "version",
+    "version_tuple",
+    "__commit_id__",
+    "commit_id",
+]
+
+TYPE_CHECKING = False
+if TYPE_CHECKING:
+    from typing import Tuple
+    from typing import Union
+
+    VERSION_TUPLE = Tuple[Union[int, str], ...]
+    COMMIT_ID = Union[str, None]
+else:
+    VERSION_TUPLE = object
+    COMMIT_ID = object
+
+version: str
+__version__: str
+__version_tuple__: VERSION_TUPLE
+version_tuple: VERSION_TUPLE
+commit_id: COMMIT_ID
+__commit_id__: COMMIT_ID
+
+__version__ = version = '0.1.1'
+__version_tuple__ = version_tuple = (0, 1, 1)
+
+__commit_id__ = commit_id = 'g697cbb526'

tau_fibrils_yolo-0.1.1/src/tau_fibrils_yolo/boxes_utils.py

@@ -0,0 +1,111 @@
+from typing import Any, Optional
+
+import cv2
+import numpy as np
+from numpy.linalg import norm
+from scipy.ndimage import zoom
+from scipy.spatial import distance_matrix
+from sklearn.neighbors import KernelDensity
+
+
+def boxes_min_distance_filter(
+    boxes, min_dist_px=5, probas: Optional[np.ndarray] = None
+) -> Any:
+    """Returns a filter matching objects isolated by less than min_dist_px pixels."""
+    centers = boxes_centers(boxes)
+    dist_matrix = distance_matrix(centers, centers)
+    np.fill_diagonal(dist_matrix, np.inf)
+    filt = np.min(dist_matrix, axis=1) > min_dist_px
+    if probas is not None:
+        return boxes[filt], probas[filt]
+    else:
+        return boxes[filt]
+
+
+def boxes_iou_filter(
+    boxes: np.ndarray, max_iou: float = 0.0, probas: Optional[np.ndarray] = None
+) -> Any:
+    """Removes boxes based on an intersection over union threshold."""
+    keep = []
+    for i, box_i in enumerate(boxes):
+        should_keep_i = True
+        box_i = boxes[i]
+        for j in keep:
+            box_j = boxes[j]
+            intersection, _ = cv2.intersectConvexConvex(box_i, box_j)
+            area_i = cv2.contourArea(box_i)
+            area_j = cv2.contourArea(box_j)
+            union = area_i + area_j - intersection
+            iou = intersection / union
+            if iou >= max_iou:
+                should_keep_i = False
+                break
+        if should_keep_i:
+            keep.append(i)
+
+    if probas is not None:
+        return boxes[keep], probas[keep]
+    else:
+        return boxes[keep]
+
+
+def clear_border_boxes(boxes: np.ndarray, probas: Optional[np.ndarray] = None) -> Any:
+    """Removes boxes touching the image border."""
+    filt = ~(boxes < 0).any(axis=2).any(axis=1)
+    if probas is not None:
+        return boxes[filt], probas[filt]
+    else:
+        return boxes[filt]
+
+
+def boxes_centers(boxes: np.ndarray) -> np.ndarray:
+    """Returns the center coordinates of the boxes."""
+    box_cy = (np.max(boxes[..., 1], axis=1) + np.min(boxes[..., 1], axis=1)) / 2
+    box_cx = (np.max(boxes[..., 0], axis=1) + np.min(boxes[..., 0], axis=1)) / 2
+    box_centers = np.vstack((box_cx, box_cy)).T
+    return box_centers
+
+
+def boxes_kernel_density_map(boxes, image, gaussian_sigma_px=50, downscale_factor=8):
+    """Returns a kernel density image from box coordinates."""
+    grid_size_x = image.shape[0] // downscale_factor
+    grid_size_y = image.shape[1] // downscale_factor
+
+    x_grid, y_grid = np.meshgrid(
+        np.linspace(0, grid_size_y - 1, grid_size_y),
+        np.linspace(0, grid_size_x - 1, grid_size_x),
+    )
+
+    grid_points = np.vstack([y_grid.ravel(), x_grid.ravel()]).T
+
+    kde = KernelDensity(
+        bandwidth=gaussian_sigma_px, kernel="gaussian", algorithm="ball_tree"
+    )
+
+    kde.fit(boxes_centers(boxes) / downscale_factor)
+
+    density_map = np.exp(kde.score_samples(grid_points)).reshape(
+        (grid_size_x, grid_size_y)
+    )
+    density_map = zoom(density_map, zoom=downscale_factor, order=1)
+
+    return density_map
+
+
+def box_measurements(box):
+    """Returns the center, length, and width of an oriented bounding box."""
+    p0, p1, p2, p3 = box
+
+    if norm(p2 - p0) < norm(p1 - p0):
+        p0, p3, p2, p1 = box
+
+    p5 = p0 + (p1 - p0) / 2
+    p6 = p3 + (p2 - p3) / 2
+
+    width = norm(p1 - p0)
+    length = norm(p2 - p0)
+
+    diagonal = p6 - p5
+    center = p5 + diagonal / 2
+
+    return center, length, width

tau_fibrils_yolo-0.1.1/src/tau_fibrils_yolo/cli.py

@@ -0,0 +1,60 @@
+import tifffile
+from pathlib import Path
+import argparse
+import pandas as pd
+
+
+from tau_fibrils_yolo.widget import detect_fibrils
+from tau_fibrils_yolo.boxes_utils import box_measurements
+
+
+def process_input_file_predict(input_image_file):
+    image = tifffile.imread(input_image_file)
+
+    results = detect_fibrils.run(image)
+    box_results = results.read("Tau fibrils")
+    
+    boxes = box_results.data
+    probas = box_results.meta["features"]["probabilities"]
+
+    centers_x = []
+    centers_y = []
+    lengths = []
+    widths = []
+    for box in boxes:
+        center, length, width = box_measurements(box)
+        centers_x.append(center[0])
+        centers_y.append(center[1])
+        lengths.append(length)
+        widths.append(width)
+
+    df = pd.DataFrame(
+            {
+                "probability": probas,
+                "length": lengths,
+                "width": widths,
+                "center_x": centers_x,
+                "center_y": centers_y,
+            }
+        )
+
+    input_image_path = Path(input_image_file)
+    out_file_name = input_image_path.parent / f"{input_image_path.stem}_results.csv"
+
+    df.to_csv(out_file_name)
+
+    print("Saved results to ", out_file_name)
+
+
+def cli_predict_image():
+    """Command-line entry point for model inference."""
+    parser = argparse.ArgumentParser(description="Use this command to run inference.")
+    parser.add_argument(
+        "-i",
+        type=str,
+        required=True,
+        help="Input image. Must be a TIF image file.",
+    )
+    args = parser.parse_args()
+
+    process_input_file_predict(args.i)