Zellerkennung
beta-11 release
https://openreview.net/forum?id=YtgRjBw-7GJ
https://bbbc.broadinstitute.org/BBBC039 (CC0)
https://bbbc.broadinstitute.org/BBBC041 (CC BY-NC-SA 3.0)
Stellen Sie sicher, dass Sie PyTorch installiert haben.
pip install -U celldetection
pip install git+https://github.com/FZJ-INM1-BDA/celldetection.git
model = cd . fetch_model ( model_name , check_hash = True )| Modellname | Trainingsdaten | Link |
|---|---|---|
ginoro_CpnResNeXt101UNet-fbe875f1a3e5ce2c | BBBC039, BBBC038, Omnipose, Cellpose, Sartorius – Zellinstanzsegmentierung, Livecell, NeurIPS 22 CellSeg Challenge | ? |
import torch , cv2 , celldetection as cd
from skimage . data import coins
from matplotlib import pyplot as plt
# Load pretrained model
device = 'cuda' if torch . cuda . is_available () else 'cpu'
model = cd . fetch_model ( 'ginoro_CpnResNeXt101UNet-fbe875f1a3e5ce2c' , check_hash = True ). to ( device )
model . eval ()
# Load input
img = coins ()
img = cv2 . cvtColor ( img , cv2 . COLOR_GRAY2RGB )
print ( img . dtype , img . shape , ( img . min (), img . max ()))
# Run model
with torch . no_grad ():
x = cd . to_tensor ( img , transpose = True , device = device , dtype = torch . float32 )
x = x / 255 # ensure 0..1 range
x = x [ None ] # add batch dimension: Tensor[3, h, w] -> Tensor[1, 3, h, w]
y = model ( x )
# Show results for each batch item
contours = y [ 'contours' ]
for n in range ( len ( x )):
cd . imshow_row ( x [ n ], x [ n ], figsize = ( 16 , 9 ), titles = ( 'input' , 'contours' ))
cd . plot_contours ( contours [ n ])
plt . show () import celldetection as cdcd.models.CPNcd.models.CpnU22cd.models.CPNCorecd.models.CpnResUNetcd.models.CpnSlimU22cd.models.CpnWideU22cd.models.CpnResNet18FPNcd.models.CpnResNet34FPNcd.models.CpnResNet50FPNcd.models.CpnResNeXt50FPNcd.models.CpnResNet101FPNcd.models.CpnResNet152FPNcd.models.CpnResNet18UNetcd.models.CpnResNet34UNetcd.models.CpnResNet50UNetcd.models.CpnResNeXt101FPNcd.models.CpnResNeXt152FPNcd.models.CpnResNeXt50UNetcd.models.CpnResNet101UNetcd.models.CpnResNet152UNetcd.models.CpnResNeXt101UNetcd.models.CpnResNeXt152UNetcd.models.CpnWideResNet50FPNcd.models.CpnWideResNet101FPNcd.models.CpnMobileNetV3LargeFPNcd.models.CpnMobileNetV3SmallFPNSchauen Sie sich auch die Timm-Dokumentation an.
import timm
timm . list_models ( filter = '*' ) # explore available modelscd.models.CpnTimmMaNetcd.models.CpnTimmUNetcd.models.TimmEncodercd.models.TimmFPNcd.models.TimmMaNetcd.models.TimmUNet import segmentation_models_pytorch as smp
smp . encoders . get_encoder_names () # explore available models encoder = cd . models . SmpEncoder ( encoder_name = 'mit_b5' , pretrained = 'imagenet' ) Eine Liste der SMP-Encoder finden Sie in der smp Dokumentation.
cd.models.CpnSmpMaNetcd.models.CpnSmpUNetcd.models.SmpEncodercd.models.SmpFPNcd.models.SmpMaNetcd.models.SmpUNet # U-Nets are available in 2D and 3D
import celldetection as cd
model = cd . models . ResNeXt50UNet ( in_channels = 3 , out_channels = 1 , nd = 3 )cd.models.U22cd.models.U17cd.models.U12cd.models.UNetcd.models.WideU22cd.models.SlimU22cd.models.ResUNetcd.models.UNetEncodercd.models.ResNet50UNetcd.models.ResNet18UNetcd.models.ResNet34UNetcd.models.ResNet152UNetcd.models.ResNet101UNetcd.models.ResNeXt50UNetcd.models.ResNeXt152UNetcd.models.ResNeXt101UNetcd.models.WideResNet50UNetcd.models.WideResNet101UNetcd.models.MobileNetV3SmallUNetcd.models.MobileNetV3LargeUNet # Many MA-Nets are available in 2D and 3D
import celldetection as cd
encoder = cd . models . ConvNeXtSmall ( in_channels = 3 , nd = 3 )
model = cd . models . MaNet ( encoder , out_channels = 1 , nd = 3 )cd.models.MaNetcd.models.SmpMaNetcd.models.TimmMaNetcd.models.FPNcd.models.ResNet18FPNcd.models.ResNet34FPNcd.models.ResNet50FPNcd.models.ResNeXt50FPNcd.models.ResNet101FPNcd.models.ResNet152FPNcd.models.ResNeXt101FPNcd.models.ResNeXt152FPNcd.models.WideResNet50FPNcd.models.WideResNet101FPNcd.models.MobileNetV3LargeFPNcd.models.MobileNetV3SmallFPN # ConvNeXt Networks are available in 2D and 3D
import celldetection as cd
model = cd . models . ConvNeXtSmall ( in_channels = 3 , nd = 3 )cd.models.ConvNeXtcd.models.ConvNeXtTinycd.models.ConvNeXtSmallcd.models.ConvNeXtBasecd.models.ConvNeXtLarge # Residual Networks are available in 2D and 3D
import celldetection as cd
model = cd . models . ResNet50 ( in_channels = 3 , nd = 3 )cd.models.ResNet18cd.models.ResNet34cd.models.ResNet50cd.models.ResNet101cd.models.ResNet152cd.models.WideResNet50_2cd.models.ResNeXt50_32x4dcd.models.WideResNet101_2cd.models.ResNeXt101_32x8dcd.models.ResNeXt152_32x8dcd.models.MobileNetV3Largecd.models.MobileNetV3SmallFinden Sie uns auf Docker Hub: https://hub.docker.com/r/ericup/celldetection
Sie können die neueste Version von celldetection abrufen über:
docker pull ericup/celldetection:latest
docker run --rm
-v $PWD/docker/outputs:/outputs/
-v $PWD/docker/inputs/:/inputs/
-v $PWD/docker/models/:/models/
--gpus="device=0"
celldetection:latest /bin/bash -c
"python cpn_inference.py --tile_size=1024 --stride=768 --precision=32-true"
docker run --rm
-v $PWD/docker/outputs:/outputs/
-v $PWD/docker/inputs/:/inputs/
-v $PWD/docker/models/:/models/
celldetection:latest /bin/bash -c
"python cpn_inference.py --tile_size=1024 --stride=768 --precision=32-true --accelerator=cpu"
Mit diesem Befehl können Sie auch unsere Docker-Images für die Verwendung mit Apptainer (ehemals Singularity) abrufen:
apptainer pull --dir . --disable-cache docker://ericup/celldetection:latest
Finden Sie uns auf Hugging Face und laden Sie Ihre eigenen Bilder zur Segmentierung hoch: https://huggingface.co/spaces/ericup/celldetection
Es gibt auch eine API (Python und JavaScript), mit der Sie Community-GPUs (derzeit Nvidia A100) aus der Ferne nutzen können!
from gradio_client import Client
# Define inputs (local filename or URL)
inputs = 'https://raw.githubusercontent.com/scikit-image/scikit-image/main/skimage/data/coins.png'
# Set up client
client = Client ( "ericup/celldetection" )
# Predict
overlay_filename , img_filename , h5_filename , csv_filename = client . predict (
inputs , # str: Local filepath or URL of your input image
# Model name
'ginoro_CpnResNeXt101UNet-fbe875f1a3e5ce2c' ,
# Custom Score Threshold (numeric value between 0 and 1)
False , .9 , # bool: Whether to use custom setting; float: Custom setting
# Custom NMS Threshold
False , .3142 , # bool: Whether to use custom setting; float: Custom setting
# Custom Number of Sample Points
False , 128 , # bool: Whether to use custom setting; int: Custom setting
# Overlapping objects
True , # bool: Whether to allow overlapping objects
# API name (keep as is)
api_name = "/predict"
)
# Example usage: Code below only shows how to use the results
from matplotlib import pyplot as plt
import celldetection as cd
import pandas as pd
# Read results from local temporary files
img = imread ( img_filename )
overlay = imread ( overlay_filename ) # random colors per instance; transparent overlap
properties = pd . read_csv ( csv_filename )
contours , scores , label_image = cd . from_h5 ( h5_filename , 'contours' , 'scores' , 'labels' )
# Optionally display overlay
cd . imshow_row ( img , img , figsize = ( 16 , 9 ))
cd . imshow ( overlay )
plt . show ()
# Optionally display contours with text
cd . imshow_row ( img , img , figsize = ( 16 , 9 ))
cd . plot_contours ( contours , texts = [ 'score: %d%% n area: %d' % s for s in zip (( scores * 100 ). round (), properties . area )])
plt . show () import { client } from "@gradio/client" ;
const response_0 = await fetch ( "https://raw.githubusercontent.com/scikit-image/scikit-image/main/skimage/data/coins.png" ) ;
const exampleImage = await response_0 . blob ( ) ;
const app = await client ( "ericup/celldetection" ) ;
const result = await app . predict ( "/predict" , [
exampleImage , // blob: Your input image
// Model name (hosted model or URL)
"ginoro_CpnResNeXt101UNet-fbe875f1a3e5ce2c" ,
// Custom Score Threshold (numeric value between 0 and 1)
false , .9 , // bool: Whether to use custom setting; float: Custom setting
// Custom NMS Threshold
false , .3142 , // bool: Whether to use custom setting; float: Custom setting
// Custom Number of Sample Points
false , 128 , // bool: Whether to use custom setting; int: Custom setting
// Overlapping objects
true , // bool: Whether to allow overlapping objects
// API name (keep as is)
api_name = "/predict"
] ) ; Unser Napari-Plugin finden Sie hier: https://github.com/FZJ-INM1-BDA/celldetection-napari
Erfahren Sie hier mehr über Napari: https://napari.org Sie können es über pip installieren:
pip install git+https://github.com/FZJ-INM1-BDA/celldetection-napari.git
Wenn Sie diese Arbeit nützlich finden, denken Sie bitte darüber nach, einen Stern ️ zu vergeben und zu zitieren :
@article{UPSCHULTE2022102371,
title = {Contour proposal networks for biomedical instance segmentation},
journal = {Medical Image Analysis},
volume = {77},
pages = {102371},
year = {2022},
issn = {1361-8415},
doi = {https://doi.org/10.1016/j.media.2022.102371},
url = {https://www.sciencedirect.com/science/article/pii/S136184152200024X},
author = {Eric Upschulte and Stefan Harmeling and Katrin Amunts and Timo Dickscheid},
keywords = {Cell detection, Cell segmentation, Object detection, CPN},
}
