Note

Click here to download the full example code

Mosquito Segmentation Tutorial

This tutorial demonstrates how to use the CulicidaeLab library to perform mosquito segmentation on images. We'll cover:

1. Setting up the segmentation model
2. Loading and preprocessing images
3. Running segmentation
4. Visualizing results

Install the culicidaelab library if not already installed !pip install -q culicidaelab

First, let's import the necessary libraries:

import cv2
from pathlib import Path
import matplotlib.pyplot as plt
from culicidaelab import MosquitoSegmenter, MosquitoDetector
from culicidaelab import ProviderService
from culicidaelab import ModelWeightsManager
from culicidaelab import get_settings

1. Initialize Settings and Segmenter

First, we'll initialize our settings and create a MosquitoSegmenter instance:

Get settings instance

settings = get_settings()
settings.list_model_types()

Out:

['classifier', 'detector', 'segmenter']

model_config = settings.get_config("predictors.segmenter")
model_path = settings.get_model_weights_path("segmenter")
provider_service = ProviderService(settings=settings)
weights_manager = ModelWeightsManager(
    settings=settings,
    provider_service=provider_service,
)
# Initialize segmenter
segmenter = MosquitoSegmenter(settings=settings, load_model=True)

Out:

Model weights for 'segmenter' not found. Attempting to download...
Ensuring destination directory exists: /home/runner/.local/share/culicidaelab/models/weights/segmentation
Downloaded weights to: /home/runner/.local/share/culicidaelab/models/weights/segmentation/sam2.1_hiera_tiny.pt

2. Load and Preprocess Image

Now let's load a test image:

Load test image

image_path = str(Path("test_imgs") / "640px-Aedes_aegypti.jpg")
image = cv2.imread(image_path)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

3. Run Segmentation

Now we can run the segmentation model on our image:

mask = segmenter.predict(image)

4. Visualize Results

Finally, let's visualize the segmentation results overlaid on the original image:

Visualize segmentation results

segmented_image = segmenter.visualize(image, mask)

# Display results
plt.figure(figsize=(15, 5))

plt.subplot(1, 3, 1)
plt.imshow(image)
plt.axis("off")
plt.title("Original Image")

plt.subplot(1, 3, 2)
plt.imshow(mask, cmap="gray")
plt.axis("off")
plt.title("Segmentation Mask")

plt.subplot(1, 3, 3)
plt.imshow(segmented_image)
plt.axis("off")
plt.title("Segmented Image")

plt.tight_layout()
plt.show()

Segmentation Using Detection Results

The segmenter can also use detection results to improve segmentation accuracy. Here's how to combine detection and segmentation:

Initialize detector

detector = MosquitoDetector(settings=settings, load_model=True)
# Load test image
image_path = str(Path("test_imgs") / "640px-Aedes_aegypti.jpg")
image = cv2.imread(image_path)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Run detection
detections = detector.predict(image)

# Run segmentation with detection boxes
mask_with_boxes = segmenter.predict(image, detection_boxes=detections)
# Draw detections
annotated_image = detector.visualize(image, detections)

# Print detection results
print("\nDetection Results:")
for i, (x, y, w, h, conf) in enumerate(detections):
    print(
        f"Mosquito {i+1}: Confidence = {conf:.2f}, Box = (x={x:.1f}, y={y:.1f}, w={w:.1f}, h={h:.1f})",
    )
# Visualize results
segmented_image_with_boxes = segmenter.visualize(annotated_image, mask_with_boxes)

# plt.figure(figsize=(10, 10))
plt.figure(figsize=(15, 5))

plt.subplot(1, 2, 1)
plt.imshow(mask_with_boxes, cmap="gray")
plt.axis("off")
plt.title("Segmentation Mask")


plt.subplot(1, 2, 2)
plt.imshow(segmented_image_with_boxes)
plt.axis("off")
plt.title("Segmentation with Detection Boxes")

plt.tight_layout()
plt.show()

Out:

Weights file found at: /home/runner/.local/share/culicidaelab/models/weights/detection/culico-net-det-v1-nano.pt

Detection Results:

metrics = segmenter.evaluate(mask_with_boxes, input_data=image)
print(metrics)

Out:

{'iou': 1.0, 'precision': 1.0, 'recall': 1.0, 'f1': 1.0}

metrics_default = segmenter.evaluate(
    mask_with_boxes,
    mask_with_boxes,
)
print(metrics_default)

Out:

{'iou': 1.0, 'precision': 1.0, 'recall': 1.0, 'f1': 1.0}

Total running time of the script: ( 0 minutes 9.414 seconds)

Download Python source code: tutorial_part_3_mosquito_segmentation.py

Download Jupyter notebook: tutorial_part_3_mosquito_segmentation.ipynb

Gallery generated by mkdocs-gallery