How chinthanipravi/satellite-oil-spill-dashboard works
This is a single-file Streamlit dashboard for satellite SAR oil-spill inspection. An analyst supplies an image and target coordinates, selects a detection mode, and receives segmentation, patch measurements, quality metrics, charts, and a map. The sampled code shows the image-processing and reporting flow; the README documents the advertised U-Net capability, though the excerpt does not establish a separate trained-model implementation. All runtime components therefore point to app.py.
Main entry points: app.py Patch Analysis
Components 15
Dashboard Interaction
- Dashboard UI
app.py - Image Upload
app.py - GPS Selection
app.py
Image Detection
- Image Decode
app.py - Engine Selection
app.py - OpenCV Segmentation
app.py - U-Net Processing
app.py - Patch Analysis
app.py
Analysis Outputs
- Patch Measurements
app.py - Detection Metrics
app.py - Image Results
app.py - Analytics Charts
app.py - GPS Map
app.py - Inspection Table
app.py
Other components
- Marine Analyst
Connections 15
- Marine Analyst → Dashboard UI: uses
README.md - Dashboard UI → Image Upload: accepts image
app.py - Dashboard UI → GPS Selection: sets coordinates
app.py - Image Upload → Image Decode: decodes image
app.py - Image Decode → Engine Selection: provides grayscale
app.py - Engine Selection → OpenCV Segmentation: selects Classic
app.py - Engine Selection → U-Net Processing: selects U-Net
app.py - OpenCV Segmentation → Patch Analysis: returns mask
app.py - U-Net Processing → Patch Analysis: returns mask
app.py - Patch Analysis → Patch Measurements: measures regions
app.py - Patch Analysis → Detection Metrics: sets metrics
app.py - Patch Analysis → Image Results: renders overlays
app.py - Patch Measurements → Analytics Charts: plots areas
app.py - GPS Selection → GPS Map: plots coordinates
app.py - Patch Measurements → Inspection Table: displays details
app.py