Merge pull request #17 from LorenzLamm/cleanup_and_docs

clean up and docs

README.md

@@ -16,14 +16,22 @@ This [napari] plugin was generated with [Cookiecutter] using [@napari]'s [cookie
 
 ## Lasso tool
 
-This repository currently contains 4 different methods for computing a 3D mask from a polygon:
+This repository allows to draw 3D lassos, generate masks from these, and then mask out the image.
+For instructions on how to use the plugin, please refer to the [Usage instructions](./docs/Usage.md).
 
+<div style="text-align: center;">
+<img src="https://private-user-images.githubusercontent.com/34575029/400214662-88851e09-6f10-4219-9b45-6f608c3e10b6.gif?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.rUYe0XfZLvXy_5iZCXqkKvP_SkZbXfKL4owh7ot8YeU" alt="lasso_gif" width="75%" />
+</div>
+
+How it works: A polygon is drawn and a mask is generated via:
 ### Mask via rotation
 Steps:
 1. Rotate and project polygon to 2D and create a pixel mask
 2. Create a 3D mask by stacking the pixel mask along z
 3. Rotate 3D mask s.t. it is aligned with the original polygon
 
+This performed more efficiently than the other methods:
+
 ### Mask via projection
 Steps:
 1. Project all points onto the hyperplane defined by the polygon
@@ -46,14 +54,6 @@ Steps:
 3. Do that for many pixel mask, varying the z-component --> will be moved into tomogram along the polygon normal
 4. Binary closing to get rid of holes from integer conversion
 
-<!--
-Don't miss the full getting started guide to set up your new package:
-https://github.com/napari/cookiecutter-napari-plugin#getting-started
-
-and review the napari docs for plugin developers:
-https://napari.org/stable/plugins/index.html
--->
-
 ## Installation
 
 pip install .

docs/Usage.md

@@ -0,0 +1,52 @@
+## How to use the lasso plugin
+
+### 1. Preparation
+- Open the napari viewer
+- Load a 3D image (binary)
+- Load the lasso plugin
+
+### 2. Draw Lasso
+First, click the "Freehand" button in the upper right corner of the viewer. Then, draw a lasso on the image by clicking and dragging the mouse. 
+
+<div style="text-align: center;">
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213952-48f24e49-4cd0-4fa1-9f22-bfe8f9ae09bb.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.w9KCehExEcqAfsGSxw8-UVTLusFaS265ckAN9LMBtag" alt="lasso_selection" width="49%" />
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213951-bf73433d-f9f2-4f7d-b798-de2c71f3197a.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.aYU0q4jzIQwb2maGQ0yiDJCXFaW3wDqGIYoFhDETy-w" alt="lasso_draw" width="49%" />
+</div>
+
+
+### 3. Generate mask and mask out the image
+Click the "Lasso" button to generate a mask in the shape of the image you want to mask. Then click the "Mask Volume" button to mask out the image and generate a new layer with the masked image ("masked_volume").
+
+
+<div style="text-align: center;">
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213950-98242f8a-bfa7-44ed-9a44-aed49c124955.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.cgEHtdFzxq1gNZ08LLQF453ZqtzGUAPtPtD4KsymASg" alt="lasso_mask_generation" width="49%" />
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213946-8dff4d4d-c0ce-4c31-abc5-f730c69f62c8.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.HJjL9w-57SLYseQdlAuyImRCiEz6kQNFLNeszLVR3rw" alt="lasso_masked_volume" width="49%" />
+</div>
+
+
+### 4. Compute connected components
+By selecting the "masked_volume" layer and clicking the "Connected Components" button, you can compute the connected components of the masked image. The connected components will be displayed as a new layer ("connected_components").
+
+You can choose the remove small objects from the components by setting the "remove small objects" parameter to something other than 0. This will remove all connected components with a volume smaller than the specified value.
+
+You also the the option to perform morphological opening before computing the connected components. This could be useful to split components that were wrongly merged by single voxels in the initial segmentations.
+
+#### Visualization
+To look at a single connected component, you can select the number of the component you would like to visualize and click "Display Connected Components" to display the selected component. All others will be blacked out.
+
+If you would like to display all components again, you can select component number 0 and click "Display Connected Components".
+
+<div style="text-align: center;">
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213937-2e4571ee-ab4c-4c01-bcfe-df3ceb2f9a60.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.hOBHQ2oVUBADDjr2Wo5VScv-9Qcelfisyqi1al430pw" alt="lasso_compute_components" width="49%" />
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213931-17253256-258e-4861-99d5-07c32f47bc07.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.MJuCpAPPt28BnjLXanxfz18rIX9D8VVYk_MYGcrZJRo" alt="lasso_look_at_components" width="49%" />
+</div>
+
+### 5. Save out the connected components
+You can now save out the components you would like to keep by selecting the corresponding component number, specifiying a file path, and clicking the "Store Tomogram" button. This will save the selected component as a new .mrc file.
+
+Alternatively, you also also specify a directory path and select "Store All Components" to save out all components as individual .mrc files.
+
+<div style="text-align: center;">
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213927-14c8b195-f439-49c4-8d9e-6af6c80c82eb.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MzYwOTYyNzUsIm5iZiI6MTczNjA5NTk3NSwicGF0aCI6Ii8zNDU3NTAyOS80MDAyMTM5MjctMTRjOGIxOTUtZjQzOS00OWM0LThkOWUtNmFmNmM4MGM4MmViLnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNTAxMDUlMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjUwMTA1VDE2NTI1NVomWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPTFmZTJhODY2YjkwNzYxMWQ3YTM1MTQxMTFlMDZmZmVmYzFjMTNjY2YwNjc0YjgwZjdkODlmYjhhOWRjMDNhZmUmWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0In0.oypEXHl5zcB1Oq9V2jh9Cgy9gY6dlKm56giuGmksapE" alt="lasso_visualize_single_membrane" width="49%" />
+    <img src="https://private-user-images.githubusercontent.com/34575029/400213919-126cab19-3c36-4556-b674-ed79d5deaa18.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MzYwOTYyNzUsIm5iZiI6MTczNjA5NTk3NSwicGF0aCI6Ii8zNDU3NTAyOS80MDAyMTM5MTktMTI2Y2FiMTktM2MzNi00NTU2LWI2NzQtZWQ3OWQ1ZGVhYTE4LnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNTAxMDUlMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjUwMTA1VDE2NTI1NVomWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPWViYmIwMDJjNDYzM2Q2NjhiODFiNDg2ZjY0ODg4NWU3MjQ2ODRkZGYxNjlhMjFjZDFmZmYwYmZjN2VlOTFlNDkmWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0In0.-h77temIeiS21EJ4vE5h_x6aq0AchFU9aNjcy8SECIo" alt="generated_files" width="49%" />
+</div>

pyproject.toml

@@ -25,16 +25,14 @@ classifiers = [
 ]
 requires-python = ">=3.9"
 dependencies = [
-    "numpy",
     "magicgui",
+    "membrain-seg",
+    "napari-mrcfile-reader",
+    "numpy",
+    "pyqt5",
     "qtpy",
-    "scikit-image",
     "scipy",
-    "open3d",
-    "pyqt5",
-    "napari-mrcfile-reader",
-    "membrain-seg",
-    "dijkstra3d"
+    "scikit-image",
 ]
 
 [project.optional-dependencies]

src/lasso_3d/__init__.py

@@ -4,10 +4,4 @@
     Lasso3D,
 )
 
-__all__ = (
-    # "napari_get_reader",
-    "Lasso3D",
-    # "ImageThreshold",
-    # "threshold_autogenerate_widget",
-    # "threshold_magic_widget",
-)
+__all__ = ("Lasso3D",)