skimage_morpho_snakes

Morphological active contour segmentation from scikit-image library. Two methods are implemented: Morphological Geodesic Active Contour (MGAC) and Morphological Chan Vese (MCV). Users must give initial level-set as input, it can be graphics input drawn interactively (Ikomia Studio only) or binary image. Algorithm creates segmented region in a binary image.

🚀 Use with Ikomia API

1. Install Ikomia API

We strongly recommend using a virtual environment. If you're not sure where to start, we offer a tutorial here.

pip install ikomia

2. Create your workflow

from ikomia.dataprocess.workflow import Workflow
from ikomia.utils.displayIO import display

# Init your workflow
wf = Workflow()

# Set input image
wf.set_image_input(url="https://raw.githubusercontent.com/Ikomia-hub/skimage_morpho_snakes/main/images/coins.png", index=0)

# Set seed image (binary)
wf.set_image_input(url="https://raw.githubusercontent.com/Ikomia-hub/skimage_morpho_snakes/main/images/seed.png", index=1)

# Add snake algorithm
snake = wf.add_task(name="skimage_morpho_snakes", auto_connect=True)

# Adjust parameters
snake.set_parameters({
    "mgac_iterations": "500",
    "mgac_balloon": "-1.0",
})

# Run the workflow
wf.run()

# Display results
binary_output = snake.get_output(0)
original_img_output = snake.get_output(1)
display(original_img_output.get_image_with_mask(binary_output), title="Morpho snake")

☀️ Use with Ikomia Studio

Ikomia Studio offers a friendly UI with the same features as the API.

• If you haven't started using Ikomia Studio yet, download and install it from this page.

• For additional guidance on getting started with Ikomia Studio, check out this blog post.

📝 Set algorithm parameters

from ikomia.dataprocess.workflow import Workflow

# Init your workflow
wf = Workflow()

# Set input image
wf.set_image_input(url="https://raw.githubusercontent.com/Ikomia-hub/skimage_morpho_snakes/main/images/coins.png", index=0)

# Set seed image (binary)
wf.set_image_input(url="https://raw.githubusercontent.com/Ikomia-hub/skimage_morpho_snakes/main/images/seed.png", index=1)

# Add snake algorithm
snake = wf.add_task(name="skimage_morpho_snakes", auto_connect=True)

# Adjust parameters
snake.set_parameters({
    "method": "mgac",
    "mgac_amplification_contour": "Inverse gaussian gradient",
    "mgac_iterations": "100",
    "mgac_smoothing": "1",
    "mgac_threshold": "auto",
    "mgac_balloon": "0",
    "mcv_iterations": "100",
    "mcv_smoothing": "1",
    "mcv_lambda1": "1.0",
    "mcv_lambda2": "1.0",
})

# Run the workflow
wf.run()

• method (str, default="mgac"): choose either "mgac" (Morphological Geodesic Active Contour) or "mcv" (Morphological Chan Vese)
• mgac_amplification_contour (str, default="Inverse gaussian gradient"): pre-processing method. For MGAC method only.
• mgac_iterations (int, default=100): iteration count. For MGAC method only.
• mgac_smoothing (int, default=1): number of times the smoothing operator is applied per iteration. For MGAC method only.
• mgac_threshold (float, default="auto"): Areas of the image with a value smaller than this threshold will be considered borders. The evolution of the contour will stop in this areas. For MGAC method only.
• mgac_balloon (float, default=0): Balloon force to guide the contour in non-informative areas of the image, i.e., areas where the gradient of the image is too small to push the contour towards a border. A negative value will shrink the contour, while a positive value will expand the contour in these areas. Setting this to zero will disable the balloon force. For MGAC method only.
• mcv_iterations (int, default=100): iteration count. For MCV method only.
• mcv_smoothing (int, default=1): number of times the smoothing operator is applied per iteration. For MCV method only.
• mcv_lambda1 (float, default=1): Weight parameter for the outer region. If lambda1 is larger than lambda2, the outer region will contain a larger range of values than the inner region. For MCV method only.
• mcv_lambda2 (float, default=1): Weight parameter for the inner region. If lambda2 is larger than lambda1, the inner region will contain a larger range of values than the outer region. For MCV method only.

Note: parameter key and value should be in string format when added to the dictionary.

🔍 Explore algorithm outputs

Every algorithm produces specific outputs, yet they can be explored them the same way using the Ikomia API. For a more in-depth understanding of managing algorithm outputs, please refer to the documentation.

from ikomia.dataprocess.workflow import Workflow

# Init your workflow
wf = Workflow()

# Set input image
wf.set_image_input(url="https://raw.githubusercontent.com/Ikomia-hub/skimage_morpho_snakes/main/images/coins.png", index=0)

# Set seed image (binary)
wf.set_image_input(url="https://raw.githubusercontent.com/Ikomia-hub/skimage_morpho_snakes/main/images/seed.png", index=1)

# Add snake algorithm
snake = wf.add_task(name="skimage_morpho_snakes", auto_connect=True)

# Adjust parameters
snake.set_parameters({
    "mgac_iterations": "500",
    "mgac_balloon": "-1.0",
})

# Run the workflow
wf.run()

# Iterate over outputs
for output in snake.get_outputs():
    # Print information
    print(output)
    # Export it to JSON
    output.to_json()

Scikit-image morphological active contour algorithm generates 2 outputs:

1. Binary segmentation output (CImageIO)
2. Original image output (CImageIO)