wzesk/littoral_refine 🖼️🔢 → 🖼️

▶️ 5 runs 📅 Jan 2025 ⚙️ Cog 0.13.2 🔗 GitHub 📄 Paper ⚖️ License
image-segmentation image-to-image remote-sensing

About

refinement module to improve satellite derived shorelines

Example Output

Output

Example output

Performance Metrics

5.42s Prediction Time
19.35s Total Time
All Input Parameters 
{
  "img_path": "https://raw.github.com/Wzesk/littoral_refine/2b77e90e1701e9da1ecf9142724ac1007848e9d1/sample/20241211T052119_20241211T052515_T43NCE_sr.png",
  "smoothing": 2,
  "mask_filepath": "https://raw.github.com/Wzesk/littoral_refine/2b77e90e1701e9da1ecf9142724ac1007848e9d1/sample/20241211T052119_20241211T052515_T43NCE_mask.png",
  "simplification": 0.5
}
Input Parameters
img_path (required) Type: string
Path to the input image for shoreline refinement
smoothing Type: integerDefault: 2Range: 0 - 10
Factor for boundary smoothing
mask_filepath (required) Type: string
Path to the input mask image file
simplification Type: numberDefault: 0.5Range: 0 - 10
Factor for boundary simplification
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
/src/predict.py:88: UserWarning: No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.
plt.legend()
Visualization saved to /tmp/refined_shoreline_visualization.png
Version Details
Version ID
86a86582a6f0095ecb14ac420ebbc1be453b84553427520b0299baedba8258f9
Version Created
January 10, 2025
Run on Replicate →